Skip to main content

Full text of "Middle Persian Literature (1968)"

See other formats



the same author La IF, 63, 40 sqq. aad 209 sqq., Wiener Zlschr.J. d. K. Siid-u-. Ostasiens, i (1957), 
81 sqq., ZDMG, 1955, '63*, aad 1957, 362 sqq. — P., Der Fremdling im Rigveda, 1939, and 
apud Altheim (v. ad § 23). — W. P. Sckmid, IF, 64 (1958}, 1 sqq. 
I 22. Ad Y 53; H. S. Nyberg, Die Religionen das alien Iran, 1938, 15 x. 

§ 23. {a) : E. Schwyzer, APA W, 1939, No. 6. — (b) : H. H. Schaeder, ZDMG-, 1940, 401 sqq. 
— (c): cf. W. Lenxz, Yasna 28 (v. ad 5 21), 991. — In general: cf. F. Altheim, Zaralhusir* ah 
Dichter, Paideuma, iii (^949}, 357 sqq, 

§ 24- O. G. VOn Wesendonk, Die religionsgeschiehiliche Bedeatung des Yasna kaptatjhdti, rgji 
(Untersuckungen z. atlg. Religionsgeschichte, Heft 3). 

§§ 26-28. The common view on Achaemsniaa Zoroastrianisni: cf. J. Duciieshe-Goillemin, 
Ormazd et Ahritnan, igS3, *7-<t5 ; George G. Cameron (v. ad § 6), >>. 5 ; A. D. Nock, A met. J own. of 
Archaeology, 53 (1949), 272 sqq. — The year 441: S. H. Taqizabeii, Old Iranian Calendars, 1938. — 
For a fuller presentation of the hypothesis her? advanced v. I Cekshevitcii, The Avestau Hymn 
to Miihra, 1959, 13 sqq.; on »ti6ra ahura v. ibid., 44, 263. 
§ 29. Translation of all YaSts: H. Lommel, Die YaSt's des Amesla, 1927; cf. also ad §§ 18-19. 
§ 30. Metre of Yasts: W. B. Hekning, Trans. Phil. Soc, 1942, 52 sqq. The reasons for the 
conclusion summarized in t 30, will be given elsewhere. 

§ 31. Hermann G-Ontert, Ueber die ahuriscken and daeviscbm Ausdrucke im Awesia, Sitzb. 
d. Heidelberger Ak. d, W., 1924, No. 13. 

§ 35. A. Christensen, Utitdes stir Is soroastruinisms de la Peise antique, 1928; Kayanidcs, 
1932 ; Die Iranier, 214 sq. 

§ 38. Mary Boyce, see ad § 2, and in Serla Canlabrigiensia (presented to Members of the 23rd 
Internal. Congr. of Orientalists], J 954, 45 sqq. 

S 40. N; cf. ad § 19(a), and A. Waac, Nirangistan, 1941, — Aog: cf. ad § 19(a), and J, 
Docheske-Guxllemin, JAs. 3936, i, 241. — Realities described ii; the V (and to a lesser extent in 
other Av. texts): W. Geiger, Ostiranische Kuliitr, 1882; Horst Fk:ktner, Die Medizinim Avesta, 
1924; A, Kammenhuber, ZDMG 108, 1958, 299 sqq. [H. Humb icii, Zlschr, f. vgl. Sprachforsch. 
77 (1961), 99. sqq. (with K. Hoffmann, ibid. 79, 238)). 

§ 41. Magi as authors of V; A. Christensen, Bssai sur la dimvnologie iranientie, 1941, 28 sq. — 
Not mentioned in the Avesta: E. Benvemjste, Les Mages dans l-mcitn Iran, 193S {contra: H. H. 
Schaeder, OLZ, 1940, 373 sqq.). 

§ 42. W. B. Kenning, JRAS, 1942, 235 sq. 

§ 43, Reiohej-t's survey: v. ad § 19, (d). — (ch. r): A. Chri-; ten-sen, £e premier ckapitre du 
Vendidad, 1043; M. MotE, JAs. 1 951, 283 sqq. — (ch. 2): Otto JUur., Wdrter and Sachen, 1938, 
176 sqq. — (ch. 3) : F. A. Cannizzaro, II capitolo georgico deWAvc^Ca, 1913. -~ (ch. 4): H. LOders, 
SPA W, 191 7, 347 s qq- 
[Note: This article was submitted in July, 1955, and slightly revised in July, 1959.] 

Some 01 the problems touched upon in the above chapter have n: 
in JNES, XXIII, 1964, 12 sqq. by the present writer, who begs 
considerations put forward in that article. One (p. 18) is that, coiit 
appears to mention Zoroaster's 'Entities' in the Behistun iuscripti 
done if his beliefs were not Zarathus'trian (ef. § 34 above). The oths 
original Old Persian inscriptional production after Xerxes (see abov 
loss on the part of professional scribes, of familiarity with the speliin 
script, and the loss, to the adoption of the Aramaic (and I would 
and script as usual means of written communication. The later ; 
themselves to copying, with' occasional slight alterations, phrases 
their activity depended indeed on what they could copy from Da; 
fact that their^phraseology is restricted to that of Darius' low-lew 
extends to statements exclusively found in the Behistun and Na< 
texts were carved too high up the rock to be legible. 

eanwhile been trea ted more fully 
leave to draw attention to two 

rary to previous opinion, Darius 
in, as he would surely not have 
■r. (p. 20) concerns the decline of 
<?, §7). This is best. attributed to 
sccouventions-cf the Old Persian 
:iow add the Elamite) language 
.uthors of inscriptions confined 
which Darius had coined. That 
■Lus, is virtually assured by the 
1 building inscriptions; it never 
:-5-i Rustam inscriptions, which 




r. Introductory 

The term Middle Persian is used of the Persian language between c. 300 
B.C. and 950 A. C. The literary remains from this long period are com- 
paratively few, and of secular works only fugitive pieces survive. For 
this there appear to be two main causes: one the radical change in themes 
and literary fashions brought about by the Arab conquest, the other the 
fact that only during the latter part of this period, the Sasanian epoch 
(c. 224 — 652 A.C.), did Persian literature begin to evolve from an oral 
{umantg) to a written [nibesiSnig) form. Writing had been used in Persia 
from the 6th century B.C. for practical purposes (royal proclamations and 
chronicles, state and private business) ; but its use was evidently not 
extended to religious or imaginative works until the early centuries of 
the Christian era. During the Sasanian epoch the Zoroastrian holy books 
and a number of secondary religious works were committed to writing 
together with a quantity of other matter more or less connected with the 
faith, which has survived under the aegis o* the Zoroastrian church The 
Mamchaean community also preserved religious writings in Middle 
Persian. 1 Secular works of entertainment, whether in verse or prose 
appear to have continued in oral transmission until after the Arab 
conquest, and coming then gradually to suffer the neglect of fashion 
passed irrevocably into oblivion. Only those one or two have survived 
which were rendered into Arabic, or remoulded in the newly-created forms 
of later Persian literature. 

The bulk of what remains from the Sasanian period is thus more of 
religious and antiquarian than of purely literary interest; yet it is of 
great importance for the cultural history of Persia. Zoroastrian literature 
having existed for centuries as a purely oral phenomenon, retained in 
its written stage various characteristically oral types of composition 
Further, a number of individual works appear to be simply oral products 
of considera ble antiquity, which were finally, because of some religious 

■ The Manichaean writings are treated separately in the next article. 



connotation, thought worthy of record in writing. The Sasanian books 
thus preserve elements from a yet older epoch, and provide a remarkable 
bridge between the two phases of Persian composition. The transition 
in Persia to a written literature, although evidently stimulated by 
foreign influences from east and west, took place without the imposition 
of an alien culture, and was thus a slow and steady process. The Arab con- 
quest came when the evolution was well advanced. The dark centuries 
which followed obscure the later phases <jf the process, and only the 
Zoroastrian books provide material by which to trace its continuity. 

The traditional forms of oral literature were evidently common to all 
ancient Iran. The content of individual works had, however, sometimes a 
local background. Most of the books preserved by the Zoroastrians were 
written down in Pars; but their subject-matter, where it has local con- 
nections, derives demonstrably not from Persia proper, but from north- 
eastern Iran. The Karndmag iArdaHr is the only such work which appears 
to stem from genuine Old Persian traditions. Evidently, just as northern 
Europe took over with Christianity a Mediterranean culture, so Sasanian 
Persia embraced with Zoroastrianism the traditions of eastern Iran. The 
same may well have been true of late Achaemeaian Persia, the eastern 
influences being merely strengthened at this la;:er period by a new re- 
course to the sources of Zoroastrian tradition. 

It is the tenacity of this tradition which makes the history of MP. 
literature difficult to write. A chronological api>roach is krgely impos- 
sible, since a work written down in the 9th century A.C. may contain 
matter transmitted over hundreds of years; whereas a 6th-century 
composition may be original, and thus later in manner and content than 
the product 300 years its junior. Further, MP. literature has in the main 
those characteristics of oral literature, namely anonymity, community 
of style, conservatism in matter and free plagiarism, which make it 
impossible to trace individual contributions or phases of growth. The 
written element hardly becomes prominent before the 6th century, 
making then, through independent authorship, its equally characteristic 
contributions of originality in style and matter; but even here, in the 
works that survive, the authority of the religious tradition acts as a 
check on individual talent. 

The persistence of ancient elements in MP. literature receives an ac- 
cidental tribute in the term Pahlavi applied to it subsequently by the 
Persians; for this word properly means "Parthian", but came in time to 
have the general sense of "heroic, ancient", and thus was used to desig- 
nate the pre-Islamic Persian culture, which seemed so remote to Muslim 



Iran. The term was applied to the older language and literature in general, 
and not specifically to the Zoroastrian works which almost alone survive. 
There is in fact evidence that in the Sasanian period, as later, it was 
secular poetry which formed the bulk of Persian literature. This poetry 
was cultivated by a highly professional minstrel body. The Zoroastrian 
works, although one or two owe something to this minstrel-poetry, were 
the products of priests, the scholars of their age, men concerned with 
knowledge and virtue, not with entertainment. They are therefore highly 
serious, and factual within the limits of theological dogma and scholastic 
learning. All branches of study pursued were developed in connection 
with the exegesis of sacred texts; and a history of the surviving MP. 
literature may therefore properly begin with the fundamental work of the 
period, namely the MP. translation, with commentaries, of the Zoroas- 
trian holy books. 1 

2. The Sasanian- A vesta with Zand 

It is not certain when the Avesta was first written down, but it is 
generally held that it was during the Sasanian period. 2 The first attempts 

1 Two mala systems of transcription of the so-called Pahlavi script exist; by the one it is sought 

to represent early Middle Persian, by the other the MP. of the Sasanian period. The second system 
is used here. There are many small differences between the two systems in the spellings of proper 
names and book- titles. r 

2 The tradition of a written Arsacid Avesta is now rejected by most; sec, e.g., H. W. Bailey 
Zoroastrian- Problems in the Ninth-Century Books (Oxford, 1943), i 5 6 ff. : H. S. Nyberg Die 
Xetigkmen des alien Iran, deutsch von H. H. Scudh (Leipzig, 1938), 4 r6 ff. In the Manichaean 
Ktphatata (ed. H. J. Polotsky and A. Bohuc-, Stuttgart, 1040), 7-31-33 it is stated; "Zoroaster 
- - , did not write any books. But his disciples after his death remembered (his words] and wrote 
the books they read today". H. H. Schaeder, ia Mor$enUm4 Heft 38, (Leipzig, 1936) p. 8of has 
pointed out that the books referred to in this 3rd century text may well be the pseud-epigrlpha 
current in, the west under Zoroaster's name, rather than genuine Zoroastrian writings, The possi- 
bility that in fact a stimulus was given by the Manichaean scriptures to the writing of' the Avesta 
has been considered (Nyberg op. cit. 415 ff., /■ Cama Or. Inst. No. 39 (1958), 30 ff ■ M Mole 
Milanges Henri Grigoire IV (r 953 ), 289 ff.}. On the various traditions concerning the transmission 
of the Avesta (all of which differ in small points) see Bailey, op. cit. 151 ff. ; Nyberg loc. cit ■ 
M. Molb Orient 13-14, 1961, 1 if. ' ' "' 

[N.B. Standard abbreviations for names and editions of Pahlavi texts are used here in footnotes 
A key to most of these can be found, e.g., in C. Bartholomae DU Zend Handschriften der K Ref- 
und Staatsbioiiothek in Munchen (Miinchen 1915), xi ff. An attempt has been made to list in footnotes 
the editions of Middle Persian tests which have appeared since E, West's survey in Grxindriss der 
iramsdtm. Philologie (1896-1904) II 75 ff. Since then all Pahlavi MSS. in Copenhagen have been 
published m a facsimile edition: Codices Avestici et Pahlavici bibliotkecae umversitatis Hafniensis 
ed. A. Chbistensen, 1031-44. The following catalogues of Pahlavi collections have appeared in 
addition to Bartholomae's: a and ed. of E. Blochet Catalogue des manuscrits maadems . ie la 
Btblwthique Nationale (Paris, 1905); M. N. Dhuu, 'Iranian manuscripts in the Library of the 
India Office' JRAS 1912, 387-98; B. N. Dhabhar Descriptive Catatogue of *U MSS. in the First 
Da$tu.r Meherji Sana Library, Sanson (Bombay, 1923} and Descriptive Catalogue of Some MSS 
bearing on Zoroastrianism and pertaining to the Different Collections in the MuUa Feroze Library 
(Bombay, 1923); J. M. Unvala Collection of Colophons of MSS. bearing on Zoroastrianism in sovie 
Ltbrartes of Europe (Bombay, 1940).] 

Handbuch der Orien talis tik, Abt. I, Bd. IV, Abschn. 2| Lfg, 1 



were presumably made with the Pahlavi script (on the deficiencies of 
which see W. B. Henning, Handbuch I, IV \, 22 ff .) ; hut this evidently 
proved inadequate for recording holy texts in a dead language, and from 
it was evolved the beautiful and precise Avestan alphabet of 46 letters. 
This, which represents a considerable technical achievement, was probably 
in being by the reign of Xosrau I, when the canon of the 21 nasks t or 
divisions of the Avesta appears to have been finally established by a 
council presided over by the high-priest Webiabulir. 1 With the care- 
fully-forged instrument of the new alphabet, the Avestan texts were 
set down in their late Sasanian priestly pronunciation. 2 

This Avesta of the 21 nasks was a huge compilation (of which the extant 
Avesta is only a small part). 3 It was held to contain "words of all know- 
ledge" (wisp-ddndgih gowisndn, DkM, 646.7}, and comprised, as well as 
the liturgical texts which survive, works on cosmogony and eschatology, 
astronomy and natural history, law and medicine, the life of the prophet 
and the history of man, extracts from ancient myth and epic, and col- 
lections of gnomic lore. The Eastern Iranian language in which it was 
composed was known to the Sasanians only as a church-language; and 
already those parts in the older Gathic dialect wer* imperfectly under- 
stood. Much labour was accordingly spent in exegesis. During the Sasa- 
nian period the whole Avesta was translated into Middle Persian. The 
translation consists of a word-for-word rendering, often faithful even to 
the Avestan syntax. Since the word-order of the two languages is very 
different, the result is necessarily clumsy and often obscure. There are 
therefore accompanying glosses, in which the sense is rendered into a 
freer Middle Persian. The MP. translation survives for the Gdthds, the 
Yasna, a few YaSts, and Vendiddd and Nirangistdn* The latter two 

» See West SBE XVIII 297 r.. -z; Bailey op. cit. 173- vrT 

* See Bailev op. cit. 193; G. Morg£.mstierne Norsk Tidsskrift f. Sprogwdenskap XII (igto) 
Qoff.: Henning Trans. Phil. Soc, London 1942, 48. ■ . 

' For the names and contents of the 21 *^S* see West SBE XXXV LI, 4 i8- +7 [some corrections 
to readings by B, N. Dhabhar The Persian Rivayats of Mormazyar Fratnarz and. others, their 
Version with introduction and notes (Bombay 1932) 4-ftJ- On *** extant Avesta see I. Gerskevitch 
in the preceding article. T ., 

« Translations must once have existed also for all the YaSts (see Bailey, op. cit. 129; J. C. 
Tavadia Die mitteipers-ische Sprache vnd Literutvr der Zarathustrier, !ig. v. H. Jwker (Leipzig. 
IM61 30).— Since West's survey the following editions or translations have appeared of the Zano 
of extant Avestan texts: B.N. Dhabkar Zand i Khurtah Avista* (test.. Bombay, 1927, translation, 
Bombay, 1963) and Pahlavi Yarn* and Visperzi (Bombay, 1949); M. K. .D*v*s. The Pahlavi 
Version of Yasna IX (Leipzig, 1904); J- Ukvala Neryosangh's Sanskrit Versus of (h 'Horn YaSt 
{Yasna 1X-X1) with the original Avesta and its Pahlavi Version (Vicuna, 1934); M. N. Diuiaa 
"The Pahlavi Text of the Orraasd Yast" in Hoshang Mem. Vol. (Bombay, 1918), 378-91. and 
Nyaishes or Zvroastrian Litanies (New York, 1408); H. JaMaSP Vmdi.ldd, Avesta text wtth Pahlam 
translation and commentary, and glossafiel index (Bombay, 1907)5 EJ. T. Abklesaria PaJdavt 
Vendiddd (Bombay, 1949) [ed. by D. N. Kapadia, author of Glossary of Pahlavi Vendidad (Bombay, 



are works not distinguished even in the original for literary merit. For 
their MP. rendering, and indeed for the MP. translation in general, it 
is impossible to claim the smallest literary distinction, although its 
philological usefulness is considerable. The translation is accompanied by 
exegetical passages, some of which are lengthy, the product of generations 
of priestly scholars, one commenting upon the interpretation of another, 
a late redactor citing the conflicting opinions of predecessors. The 
individual commentators, many known by name, are modest, and pre- 
pared to admit perplexity; "To me it is not clear" (km ne win); but 
collectively their work had immense authority, for together with the 
interpretive glosses it made up the Zand. 

The exact meaning of the word Zand is uncertain, but it was probably 
something like "understanding" or "elucidation". There evidently existed 
an ancient Zand in the Avestan language, and some Avestan glosses have 
been incorporated in the texts they interpret; but to the later church the 
Zand came to mean above all the MP. interpretations of the holy texts, 
whether glosses or exposition. As the heir (in theoiy at least) to an ancient 
exegetical tradition, the MP. Zand was accorded great authority, and 
even held to be divinely inspired; 1 and eminent Sasanian priests, no less 
than Zoroaster's first disciples, were venerated as poryotkesdn "possessors 
of the primitive [i.e. the true] faith". 

No clear distinction is made by the Phi. writers between the Avesta in 
the Avestan language (ewdz i abastdg nam, DkM. 455.11) and its MP. 
translation. 2 In the MSS. the translation, with the Zand, follows the 
Avestan text verse by verse or sentence by sentence; and both text and 
commentary were memorised. There were learned priests said to know 
the whole Avesta with Zand by heart [hamag abastdg ud zand warm, 
Ep.M.14.11); and stories are told {Dk. VI) of priests reciting together 
the abastdg ud zand as they went about their other labours. As late as 
the 9th century it was explicitly stated: "it is reasonable to consider the 
living spoken word more important than the written" (zindag gowiSmg 
saxwan &z an i pad *nibist mddagwardar hangar dan cimig). 9 It does not 

1953)]; S. J. Bulsaha Airpatasidn and Nirang&stdn ... translated (Bombay, 1915); A Waag 

Xirangistan (Leipzig, 1941) [Avestan text, with Pahlavi translation and glosses, but without the 

Pahlavi commentaries]. M. F. Kakga has translated the Pahlavi Version of YaJts (Bombay 104.0 • 

the Pahlavi Version of Afrinagdn ArUkfmvakhsk (Scmj VarUtman Annual, Bombay, 104.7 no r-*| 

and Pahlavi Ya S n Ha V 111 (Siddha Bhtira.ll, Hoshiarpur, 1950, pp. 1-7). For penetrating genera! 

comments on the Phi. translation see K. F. Gexdner, GIF. //x§§ 4^-49 (English translation by 

D. Mackichan m Avesta, Pahlavi and Ancient Persian studies in ho>iour of ...P. Saltan a (Strass- 

burg and Leipzig-, 1904), 71-81). ' 

> See J. Darmesteter tildes itaniennes II 43 a . 2; Bailev op. cit. 162-63 (citing DkM 4<?o 8 f f ) 

1 See Bailey op. cit. 167. " '' 

' J DkM. 460. 7-8 (see Bailhv op. cit. 163). 



seem likely that many MSS. of the whole Avesta ever existed; and pro- 
bably for this reason the Muslims did not consider the Zoroastrians as 
"people of the book." 1 Nevertheless, for the most detailed sections of the 
surviving Zand, those of the Vendiddd and Nnangistdn, not completed 
till after 632 A. C., 2 it is difficult not to suppose that it was the existence 
of the written text which made possible the elaboration of commentaries. 
These two works are largely concerned with matters of ritual and ob- 
servance, and the commentaries therefore remain narrowly within 
ecclesiastic tradition. It is evident, however, that in the Zand of some 
of the lost nasks the priestly scholars drew from wider fields of foreign 
science, making use, that is to say, of foreign books. The evidence for 
this comes from two sources: the later Pahlavi compilations, derived 
from the Avesta, and old Arabo-Persian works. Thus, for example, 

"there is much evidence in the old [Arabic] books of astronomy 

and astrology to show that the astronomical knowledge obtained from 
old Persian sources was mostly derived from the Sasanian Avesta 
itself." 3 The chief foreign influences to be traced in MP. science and 
literature were those of India, Greece and Babylon. Some Greek in- 
fluence is perhaps ancient, but Greek writings were presumably known 
to the Sasanians through Syriac translations from the 5th century on- 
wards. The following Greek and Indian learned works have been traced 
in Middle Persian writings, or in Arabic translations from them: 4 from 
Greece (probably largely through Syriac translations) works by Aristotle, 
including his Tcepl ye-viasat x<xl 98op6ce, and by non -Aristotelian philo- 
sophers, with general references to "Greek philosophers" [hvom filisojdy) 
and the Sophists (so/istdy); Ptolemy's Almagest (magistag I hromdy by 
Ptalamayus) and other mathematical works, including books on geome- 
try (nibeg % zatmg-paymanth) ', the Hippocratic treatises rapt cpucrioc; 
avQpuTrou and 7cspi; epSofiackov 5 and other medical works; astronomical/ 
astrological works 6 by Dorotheos of Sidon, Teucros and Vettius Valens, 
as well as the legendary Hermes; and works on agriculture. Greek moral 

1 S«e BArtEY op. cit. 1.69 ; Nybebg Die Retigionen 13, On the Syriac evidence for oral transmission 
see F. Nau RHR XCV 1927, 149-99 and J A 1927, *5o "-; J. de Menasck BSOS IX 1958, 587 a. 2. 

* See WKKC'NiTangisidn 10. 

3 S. H. Tao,izadeh BSOS IX L (1937), 135- 

* For the first-mentioned works see BaIlky op. cit. 80-87, 98. 105; for some supporting evidence 
on the works of Aristotle see R, C, Zaehnex Zurvan (Oxford, 1955) (index s.v. Aristotle). 

6 The theory that this treatise represented a pre •Sasanian, influence- from Iran on Greece 
(A. Goize Z1I II (1923), 167 fl) has been disproved; see J. Duchbsne-Guillemis Harvard 
Theological Rcjiew XL1X 2 (1956), 115 if.. The Western Response to Zoroaster (Oxford, 1958), 72 ft'. 

* On these see C. A. Nau-ino in A Volume of Oriental Studies presented to E. G. Browne (Cam- 
bridge, T922), 345 ff., and in. ERE XII 90-91; A. Barissoh J A re.35 i, 300-05; S. H. Taqizadeh 
BSOS IX i (1937), 125 ff. 



philosophy does not appear to have been studied. From India * came 
works of logic (far* < Skt. tarka) and rhetoric [avyakaran < vydkarana) 
treatises on astronomy (making known the zig f kindug, and the Indian 
system of lunar mansions) and on horoscopes (kora); an d a compendium 
on the origin of the four castes from the primordial Purusa * Indian me- 
dicine as well as Greek, was studied at Weh-andyok-sabuhr (Gundesabur)* 
and Indian herb-names are found in the Pahlavi books.* Babylonian 
astronomical works were also probably translated during the Sasanian 

The history of Sasanian astronomy provides some help for datine 
these translations. "Under Sassanian rule there were two periods of 
contact with Greek and Indian science during which the study of as- 
tronomy was promoted: one under Shapur I after the conclusion of the 
Roman war, the other 'towards the end of the Sassanian period' " i 
Phe Persian astronomical canon {zig l sahriydrdn) was drawn up during 
the first period, in 263/4 A.C. ; • the Indian system of lunar mansions waf 
borrowed during the second, ± 500 A.C It seems likely that these two 
periods were favourable for intellectual activity and foreign contacts 
mineral. Such works as have an indication of date belong not sur- 
prisingly, to the later period. 9 

The translated works were evidently preserved as independent treati- 
ses some being in due course rendered into Aramaic or Arabic. They 
probably also inspired independent treatises by Sasanian scholars (one 
Andarzgar is mentioned as an astronomer who wrote books). A tradition 
exists however, that these scientific works from abroad were also some- 
how brought into association with the Avesta itself. The MP phrase i« 
aoagabaOig abdz handdxt (DkM. 412.21). The precise significance of 

' For these sec Bailey op. cit. 80-81, 86; de Muniscb 14 m^ r,„ „ ., „ ~ 

»*«- *" **•*-< (^ris, ,958), 27; H«.n. B "Sli 9 2;?r 3 ' UKe ****** 
Sec S. K. Hodivala 1,1 Oriental Studies i» honour of C 'e Pamv ff ,«,*, \ 

s Oa the traces of Indian medical theory in Sasanian works w TV vh, ,^„ b - ,. • 
m*» Bucte KalUa wa Di,nn« KerseJt wld ^Sb^X f It .^ 
L'Tran s»w Us Sassanidcs (2nd cd., Copenhagen 1044) 435 Ciihct«ks«x 

* Bailey op. cit. 81 withn. 3. 
5 Hekning JRAS 1942, 245. 

* Taqizadeh op, cit. 134 ff. 
' HENNrS'O loc. Cit. 

* There is a Manichaean test which sUL'gcsts that Sabuhr\ s«„ u,i„.- T 

with the study of medicine ( See H ■«,« fjJSOS X 4 SS 0< n iwt" ' ** *"* "' ^^ 
of a geaeral reaction against foreign science dwfog ti S£n >* lllCrCaSC5 thC likc,lhood 

si'ssr {0n literature and scieL in <™^®^^ 



abdz handdxt is debated (the words have been variously translated as 
"restored to", "joined with", "compared with" etc.); 1 but whatever 
the shade of meaning, the evidence of the Pahlavi compilations and 
Arabo-Persian books shows that foreign learning did somehow penetrate 
the holy writings; and since it is improbable that any but an imitative 
and stammering Avestan was composed in Sasanian times, it must have 
been the MP. text that was thereby affected. The MP. translators of the 
Avesta aimed, however, at complete faithfulness, and therefore it can 
only be in the Zand that new material was adduced. That this in fact 
happened appears from the derivative Phi. works, where ancient dogma 
and later science subsist side by side. Thus the Avestan texts evidently 
preserved such pre-Achaememan beliefs as that, for example, the sun 
and moon were further from the earth than the stars; 2 it was the com- 
mentators presumably who, helpless to alter dogma, adduced nevertheless 
more advanced Babylonian theories, and Greek and Indian teachings, 
thus incorporating these in the "Avesta with Zand", and giving them, 
in course of time, the authority of "sayings of the ancients" (poryoikesdn 
gSumSrum). To the commentators also must be credited such difficult 
fusions as the introduction of the four Empedoclean elements into 
traditional cosmological doctrines, * and the association of Hippocratic 
and Indian teachings with ancient Iranian medical theory. 4 The sur- 
viving Zand, suggesting as it does that the Sasanian priests were en- 
grossed by a narrow study of ritual and ordinances, does them injustice. 
Their work had evidently a far greater breadth; and to their intellectual 
labours * Islamic Persia was to owe much. 

The assimilation of foreign learning depended on the use of books. 
Thus the exegesis of the Avesta thrust Sasanian scholars into book- 
learning, and must have been a powerful stimulus to the growth in 
Persia of literacy in the narrow sense. 

3. The later religious Writings 

The Avesta being such a huge and miscellaneous compilation, it was 
natural that smaller collections of holy texts should i:xist separately. The 
liturgical texts which form the extant Avesta were evidently in being as a 

1 See most recently M. Mole Oriens 13-14 (1961), 5 it- 

* See Hennikg JRAS 1942, 229 ff. 

3 See Zaehner Zvrva,n 141. 

* See ChriSteusen L'Iran 418 ff. 

■ * On -'the severity of the intellectual tasks confronting; the Sasanian Driestly scholars see further 
Bat ley op. cit, 117 ff. 



group in pre-Sasanian times. The Nnangistan also survives separately. 
with part of the Herbadistdn, as a work of practical use for priests. It too 
is accompanied by its Zand. 

A step which must, however, have been made after the establishment 
of the canon was that of excerpting from the Avesta passages relating 
to particular themes. This is an activity belonging necessarily to an es- 
tablished written tradition, whereby it becomes possible to select, compare 
and compile from texts simultaneously available. When the first of such 
selective compilations was made is unknown; but in those which survive 
only the MP. translation with Zand is drawn upon, the passages being 
chosen and set together to form a continuous treatment. This develop- 
ment, which seems so natural, as the comprehension of Avestan dwindled, 
may well in fact have been a slow and difficult process, since the holy 
language was deeply venerated. There is no evidence for such compila- 
tions before the Arab conquest; and although an argument ex silmtio 
is dangerous in this sparsely-documented field, it is nevertheless probable 
that the development took place slowly during the 7th to 9th centuries 
being stimulated perhaps by the needs of a church on the defensive' 
Freed from the thorniness of texts in a dead language, such compilations 
must have been both easier for Zoroastrians themselves to study, and 
more persuasive for those wavering in the faith. 

A good illustration of the development is provided by the short 
treatise Sdyisl ne-sdyist ("Allowed and not-allowed"),* which covers 
a number of miscellaneous topics, loosely joined by the thread of sin 
and ritual purification and atonement. Its matter derives directly from 
the Zand. A number of nasks are cited, mostly from the legal group - 
but it is more often the commentary than the translation which is in- 
voked, sometimes in general terms, "the authorities have taught" 
{dastwamn cast), often with the name of the commentator.* Similarities 
in style suggest that Sdyist ne-sdyist was compiled by the same priests 
who made the final redaction of the Zand of the VendUdd and Niran- 
g%sta.n (evid ently, rituaUy-minded men).* It must accordingly have been 

1 A " e f tfon of *'! * ex > r witil supplementary texts, by M. E. Davar was priri ted in Bombay in 
1912, but never published It „« destroyed by fire at the Fort Printing Pre**, Bombay Sr 945 
A few copies are in private hands, and one of these was used {with acknowledgements] by TC 
Tavad.a m h:s edition Styest ne-tayast (Hamburg, i 930 ), which has th & texf^rLcription 
only, together vn h English translation, notes and glossary. A similar edition of the supplemental 

s^tit ted for L doc , torai thests of Boiabay universit y * p - m. Kor WA , S.-S 

2S?J!?S *7 *k , ifT" 1 -- U$Ually thCy are "Rafter the * °P«4 words, or" 
significant phrase (as here), or by a descriptive title §iven later 

' For the names of the commentators cited see Barthoiomab Zend Hss. 49 = two separate 
priestly tradihons have been discerned among them (Tay.uma op. cit 20 n ! 
1 See Darmestkter Et. ir. II 4.2; Waag op. cit. 4.4. 



written after 632 A.C. Although its casuistry is -unattractive, the treatise 
is interesting for its religious technicalities and rare terms. 

A much more important and fundamental work of compilation is the 
Bundahisn ("Creation"), also called Zand-dgahih ("Knowledge from the 
Zand"), 1 which survives in two recensions, the Great (or Iranian) Bun- 
dahisn and a shortened version, the Indian BunJahiSn (deriving from a 
different MS. tradition). 2 One of the two great Zoroastrian compilations, 
this work probably grew through different redactions, from some time 
after the Arab conquest down to 1178 A.C. (whtn a few additions were 
made in imperfect Middle Persian). The last important redaction belongs 
to about the end of the 9th century. 3 The BuvdahiSn has three main 
themes: creation, the nature of earthly creatures, and the Kayanians 
(their lineage and abodes, and the vicissitudes befalling their realm of 
Eransahr). The compiler does not name individual sources; but the work 
shows an encyclopaedic knowledge of the Zand/ and exemplifies ex- 
cellently the process whereby treatises on chosen themes were created 
out of the scriptures. Many passages evidently derive fairly closely from 
the Middle Persian translation, for an Avestan syntax underlies them; 

1 Against Chris tense n's rejection of this second title {Kayanid.s 45 f.) secTiWADiA op. cit. 74. 

1 See West GIP ii 98. The Great Bundahiin, ed. by T. D. Anklesahia, was pub. by E. T. 
Anklesaiua (Bombay, 1908)- For this edition tbe MS. TD* was u-«d. TD 1 , a -very good MS., was 
not collated, A copy of TD 1 , made for Dasmesteter, is in the Bibliotheque Nationale, Paris (MS. 
Supp. Pars. 204 3), This was collated by H. W. Bailey, who kindly made his collation available 
to other scholars (see Chrtstensen Le premier ckapilre in Vd., 8; Zaehner Zwvan, 377), A 
complete English translation with notes was made by Baji-ey for a D.Phil thesis in Oxford in 1936. 
Another English translation, under the title Zand-akasili, was madu by B. T. Ankiesaria, but the 
printed copies perished in the fire of '1945; it has now been published by offset process by the 
Rahnumae Mazdayasnao. Sabha (Bombay, 1956). An edition by X. Bark with text, translation, 
notes, glossary and concordance, is in preparation. The following chapters have been separately 
edited since West's survey; I and III by NVbero J A X999 i, 207-37. 260-310, and by Zaekner 
Zurvan 278-33$; I also in Nyberg's Hiljsbwh its Peklevi (Uppsala, 1928) [on GBd. 10 see Bailey 
Zcr. Problems 94 n. 2 with references]; II by Kenning JRAS 1942, 225-48; tbe beginning of IV 
by Zaehner op. cit. 355-9 (see Christeksen Le Premier Hom-me (Uppsala, 1918) 15 ff., E. Benve- 
niste MO XXVI (1932}, 1B711.]; extracts from V, and XXV, by NyBERG TexU zum rmzdayas- 
■niscken KaUnder (Uppsala, 1934), 10-29; XXVIII by A. Gotze ZI1 Bd. II (1923). 60 ff.; XXX by 
J. J. Momin A -paper read before the Bombay branthofthe HAS, 1901 (pub. Bombay, 1902); XXXIII- 
XXXIV by 0. Messina O/imtalia IV (1935), 259 ff- A series of passages are transcribed and trans- 
lated by Schaedee in Studien z-um anlikett Synkretismus 213 ff.% £«d by Zaehner in Zurvan (see 
index, p. 4Q0); GBd-. p. 51 ff. is translated with notes by S. H. TAftrzACEH, Gah-Sumdri dar irdn-i 
qadim, "(Tehran, 193.*), 326-2.9, see further D. N. Mackenzie, BSOAS XXVII 3 (1964), 5"-29- A 
number of passages were transcribed and translated by M. Mole in bis posthumous Cvlte, Mythe 
et Gosmologvs-da-ns I'Iran antien (Paris, 1963), see index, p. 592. 

3 See GBd. 237.15 ff, The passage (upon the genealogies of priestJy families) is not found in the 
Ind. Bd. Unfortunately it is not wholly clear, either in the forms of some names or in the degrees 
of descent. On it see Modi A Paper read before the Bombay branch of Ike P. A S , 19&1 (Bombay, 1902), 
6; BsABHAft NM. p..j n. r; B. T. Anklesaria GBd. 11; Tavadia'op- cit. 75- On corroborative 
evidence for a late gti-century dating see West SBE V xli-xiiii; Parmesteter in a paper read to 
the JarthJ?shti-Dinm J fcho2-karn4ri MandU (Bombay, 1887). 

« See Christensen Kayanidss 47 f.; HENNrNG fRAS 1942, 229. 



and one section consists simply of the translation of the ist chapter of the 
VendUad coinciding (except in small details) with the canonical Zand.* 
Glosses and commentaries provide part of the continuous text, and in 
these, foreign learning is adduced. There are also a few isolated attempts 
to bring the work u P to date, by the identification of traditional {and 
even mythical) geographical names with Arabic ones. In the main 
however, the absorbing interest of the Bd. lies in the antiquity of its 
material. Here is preserved an ancient, in part pre-Zo roast rian picture 
of the world, conceived as saucer-shaped, with its rim one great mountain- 
range, a central peak thrusting up, star-encircled, to cut off the light 
of the sun by night; a world girdled by two great rivers, from which all 
other waters flow; in which yearly the gods fight against demons to end 
drought and famine, and to bring protection to man. Natural phenomena 
are speculatively explained; the sprouting of the plants, for example 
is ascribed to the mythical Tree of All Seeds growing in the ocean, whose 
seeds are mingled with water and so scattered annually over all the 
earth when the god Tistar brings the rains. Not only is the matter an- 
cient and often poetic, but the manner of presentation, although arid is 
of great antiquarian interest; for after the distinctively Zoroastrian 
account of creation, the speculative learning and legendary history is 
set out m traditional oral fashion, that is to say, in schematised mnemonic 
lists: so many types of animals, so many kinds of liquid, so many names 
of mountains, so many great battles. This is the learning of ancient Iran 
as it must have been evolved and transmitted by generations in the 
priestly schools, 

If the Bd. is compared with, for example, such a work as the Hudud 
al-Alam, so little later than its final redaction, it becomes probable that 
one reason why Zoroastrianism yielded generally to Islam is that over 
the centuries Zoroaster's noble religious and ethical conceptions had 
become fused with a mass of traditional lore, which acquired through 
association with his teachings the authority of dogma. The later adherents 
of Zoroastnanism were thus straitened by outworn conceptions, and 
many a Sasanian must have been torn between orthodoxy and scientific 
truth. The new religion offered escape to a relative freedom of the mind 
The Bd. has in the main the anonymity of traditional material ; but the 
gth-century redactor gives his own genealogy, and names some of his 
contemporaries, among them Zadspram. son of Juwanjam, selections 
from whose own writings also survive. These Selections, the Wizidagihd I 

' See Christensen Lt premier chapitre du Vd., 7. 

4 2 


Zadspram, 1 likewise consist of excerpts from the Avesta and Zand, on 
several themes: creation, legends about Zoroaster, the formation of 
man (from body, life and soul), the deeds of the hero-prophets, and the 
restoration of the world of good. The work shows the advantage of a 
single hand. It is much slighter than the Bd. ; but where the two coincide, 
the Wizidagika. are noteably free from the repetitions which characterise 
the other work (created presumably by successive redactors adding 
fresh matter from different parts of the scriptures). No Pahlavi work is 
easy reading, because of the script, and the obscurity of allusions to 
events and persons since forgotten; but these general difficulties apart, 
the Selections of Zadspram provide in the main a plain and intelligent 
summary of some of the fundamental legends and beliefs of the Sasanian 
church. In places, e.g. on the formation of man, there is a noticeable 
tincture of Greek philosophy, derived presumably from the exegetical 
passages of the Zand. 2 In these more complex passages the difficulties of 
the text can be considerable. 

Zadspram also wrote a "Compendium on the Enumeration of Species" 
(loxm-usmdrismh hangar dig nib eg)? unfortunately lost. Evidently his 
interests lay in fields other than ritual and observance; and in fact he 
attempted -to introduce a simplification into the important baresnom 
purification-ceremony, which aroused indignation among his congregation 
at Sirkan. They appealed to the high-priest of Pars and Kirman, who 
was Manuscihr, Zadspram's own brother. Three letters from Manuscihr 
upon this subject survive, the Ndmagiha I Manuscihr:* one to the con- 

J See West GIF ii 105; an edition by West of the early chapters, with transcription, was 
subsequently published as Appendix II to Avesta, Pahlavi and Ancumt Persian Studies in honour 
of ... P. Sanjana {Strassburg and Leipzig, 1904). An edition by B. T. Anxlesaria was largely 
destroyed by fire at the Fort Printing Press in 1045. It was published in 1964 by the Trustees of 
the Parsi Panchayat in Bombay as the Vtchilakxha-i, Zadsparam, Part I, text with introduction, 
which includes 63 pages of a summary of the contents of the work, by Anklesaria. The text has 
been set up exactly according to the original edition, but in new type. The following passages have 
been published, mostly from surviving copies of AnklEsama's original edition: Ch. I and XXXIV 
transcribed and translated by Zaehner BSOS IX 3 (193S), 573 ff-, X 2 (1940) 377 f f, (see also his 
Zurvan 339 ff.}; Ct. XXIX-XXX transliterated by Bailey Zor. Problems 209-16. Many shorter 
passages have been transcribed and translated by various scholars, notably by Schaeder in 
R. Rbitzenstein and H. H. Schaeder Studicn turn a-ntikm Synkretitmus aut Iran- und Griecken- 
land (Leipzig, 1926), 213 ff,; by Bailey -op. cit. and Zaehmer op. cit. (see index, p. 461); Mole 
CulU . . : (see index, 591-92). There is a translation of part of the text among the late M. Mole's 
papers, the publication of which is expected. 

s See bailey op. cit., 104 if. It seems unlikely that, with the church impoverished and on the 
defensive, Zadspram himself should have had either opportunity or desire for direct recourse to 
translations from the Greek. 

3 See Wizidagika IX 16 {West SEE V 181). 

* See West GIP ii 104; ed. (with English intro.) by Dhabhar The Episites of Manushchihar 
(Bombay, 1912)^ M. F. Kakga has transcribed and translated Ep. II i in Deccan Bulletin XVIII 
(1958), 374-80, and translated Ep. Ill in Povrt Davoud Mem, Vol. [I {Bombay, 1951), 189-204, 



gregation, one to Zadspram, and an open one. The last is dated to 881 A 
C. These letters are dignified, firm, and strongly traditionalist. Manuscihr 
seeks support unquestioning^ from the church fathers in upholding 
established practice. He awakens respect by his seriousness and concern 
for the spintual welfare of his people, but unfortunately as a stylist 
leaves much to be desired. His meaning tends to be clouded by an un- 
necessary profusion of words, and clogged by the weight of elaborate 
periods and compound terms. This is the more unfortunate as he has left 
also aa important work, the Dddisidn I denig "Religious Judgments" 1 
consisting of his answers to 92 questions put to him by lay co-religionists 
on subjects ranging from doctrinal matters to ethics, cosmology, the rights 
and duties of the priesthood, and social and legal points (adoption 
inheritance, etc). Some of the practical problems are new, those of a 
community of waning wealth and power- most can be answered autho- 
ratively from scripture. A large amount of the subject-matter is thus 
traditional, as is the form (question and answer, a common type of oral 
composition). Manuscihr>s works nevertheless belong much more to 
the 9 th century than do the previous books discussed, which appear 
in content almost wholly Sasanian. 

A gth-century element is also strong in the Denkard, "Acts of the Re- 
Ilglon "' 2 the most massive oi extant Phi. writings. This is an encyclo- 

1 Sec West GIP ii 102, Part I (Pursisn I-vr \ „a\^a k„ t t^ a 

B. T. Ankle S ar,a (Bombay, date?); ParVlI (itf XU Xr^n ^T' "** ^^^ by 

as a doctoral thesis in London u. nversit - i u i \ 8 Tn fnLw '' ^ "^J* R K " A * K «*«" 
and translated into English: Purs Un,"v$x £ D^Zf T % ^ ^ ^ ttanscribed 
Franwz and others, (Bombay. Ig32 ), x8a '. 8 «,. p^' £ c h v Tt \ " ^ ° f Hwt>w ^ 

Me,n. Vol. (Bombay 1943). 232 L ' ^ "{ * "J f^ 1 - ^k» wiha m Dinskaw j. Irani 

Voluw (Bombay, .964), :2 7 - 4 o. In tte SrrTvotoS i K ^ Y AN ° A " the DK Unvaia Mcm ' 
criptbn and translation of Pur S XCI 5Tv ,h™ T n I ^"^^ ? ublish ^ the text, trans- 
- translated by EuropLf Solars i^ff^^^ T? f—ibed 
p. 459), »nd M. Mole, op. cit. {see index, p. 593) * ° P " ^ ******* <*■ «*• (** '«d S x, 

2 See West G/Pii 91. The edition bcffunbv P R S**,.™ ;« T . . 1 

The two-volumed edition of D. M K Twill f toi {££ f 9 ^^4, wasiinis ^ d « »9»«. 
details of these editions, aad of ^y^JSL^^JS'T'i '" T^ i " ^ F ° T 
clopUi* maidtcnne, Le DMurt, Quail S^lTd^lf^u I « Me««« Une Eny. 

Cama Inst.) has since been published" y Jfj d£^ SJT f "*■ * ^ the K " R - 
satisfactory translation exists of any of the 6 hook £ S SSw ZT "■ *?■' Tf ^^ n ° 
been transcribed and translated, notably by B/Sw« J? T^t^r ^ ^ ^^ haVe 
2,nnm, (see index, p. 459j of his Zt^)^S^\^^l^^^ K f^ r) ' 
in 1946), de Menasce gave a detailed aaalysis Q s the contend 5 ^.^^/^^^dehvered 
of individual passages, and a tabte of ei^ft"?^^ ^ .^T 
edition and translation of Bfc. VII by M Mole nnw to ™~, ^ .t }' He ann0UJlced an 

45-73) also devoted considerable attention *££ZmV%%F^^ r l? Arm (QP " Cit 
as a in Paris in r 9 6 2 by B. Faravach \nd tZtZ a' I ^ transcnbed and translated 
thesis by S.. Shake, in%6 4 \ \ I Z^m£ SSSS?" 7 S aft? * * M » 
from the D*W (see inde X , pp. 5 8 9 ., l); „/. ^^ HI JSadan pt T^STZ 



paedic work of great length, which survives in mutilated form, compris- 
ing 6 books (numbered 3-9; the beginning of the 3rd book is missing). 
The Denkard, like the BundahiSn, is a compilation, which evidently- 
absorbed older works as it grew; 1 but in its final form it is attributed 
to two authors. One, Adurfarnbag i Farroxzadan, was high-priest of Pars 
at the time of Caliph Ma'mun {813-33 A.C.). 2 The first part of the Den- 
kard, made up, it seems, of selections from his writings, suffered damage 
after his death. It came eventually into the hand? of yet another high- 
priest of Pars, one Adurbad i Emedan. His date is uncertain, but it seems 
likely that he is to be identified with the Adurbad i Emedan whose son 
Isfandiyar (likewise high -priest) died in 936 A,C. S He himself should 
accordingly have flourished at about the turn of the century, which makes 
his work roughly contemporary with the main redaction of the Bun- 
dahisn. He is therefore presumably to be further identified with the 
Adurbad i Emedan who is mentioned in the BundahU-n^ as a contempo- 
rary of Zadspram's. Adurbad restored the work of Adurfarnbag, and 
added to it the later books. 

The Denkard is a vast treasure-house of Zoroastrian knowledge, 
wherein materials from many epochs and varied origins are rehandled at 
a late date by individual authors. What makes it the most formidable of 
MP. works is the style in which the greater part is written, tortuous, 
intricate and dry. Since to the inherent difficulties of this style are added 
those of the script and of a poor MS. -tradition, much remains obscure. 
The Denkard is far too long and varied a compilation to be briefly sum- 
marised; much of the first part is apologetic, and most of the second 
devoted to preserving knowledge of the Zoroastrian faith. In both, much 
material is attributed to the "exposition of the Good Religion" (nigez 
I wehden), probably the written Zand reinforced by contemporary oral 

transcribed and translated by him in the Dr. U-nvala- Memorial Vol. (liombay, 1964), 35-39. In tho 
same work, pp. 99-312, H. S. Nyberg treated the opening of Bk. V. In Indian Linguistics XXV 
(1964/5), PP- 3-20, M. F. Kakga gave a transcription and translation rf a long passage from Bk. V 
(Madan pp. 454,21-470). 

1 It is said, e.g., to- nave included the work of one Adurbad 1 "Jaw.uidan {SGV. IV 106). 

* There is a small Pahlavi treatise, G-uzastag AbdliS [ed. A. Bakthelemy, Paris 1887; H. 1\ 
Chacka, Bombay 1936] relating Adurfambag's dispute, in the presence of Ma'mun, with an, apostate 
Zoroastrian, 'Abalis. [This name, evidently a corruption of an Arabi-: one, is perhaps to be read 
Aba Laif), see Sckaeder Iranischt Beiir&ge I {Halle, r»3o) 287 0. 3.] 

s See Mas c udi KiiSb ut-Tanbih wa 'l-Askr&f, transl. B. Carra de Vaux (Paris, 1897), p. 149 
with a. 1. More recently several scholars (J. J. Modi in Studio. IndO'Ir.inica, Ehrengabe f. W- Geiger 
(Leipzig, 1931), 274-88; de Menasce op. cit. 10-n; and Tavadia op, cit. 50) have agreed in 
identifying' Adurbad rather with the Son of Emed 5 ASawahiStan, high-priest in 955 A.C. This 
would put bis redaction of the Dinkari in the second part of the 10 ih century, which seems too 
late a'* date for such a massive Middle Persian compilation.; but see further de Menasce 
A nnuaire de I'icoie pratique des fazitUs Sludes, Section des sciences religieuses {Paris}, 1956-7, pp. 9- 1 1 . 

*■ See GBd. 238.3- 



tradition.* In the early books are to be found answers. to questions by 
heretics, and by a disciple; collections of precepts [andarz) used apolo- 
getically (heretical utterances and their orthodox parallels)- an abrid- 
gement of the answers given by Adurfarnbag to a certain Ya<qub, and 
to a Christian, Boxt-Mare. To Ya'qub Adurfarnbag uses appeals to 
national pride rather than theological arguments, linking the glorious 
history of Iran with that of the Zoroastrian faith." These early books also 
contain such well-known passages as a long chapter on medicine,* and the 
accounts already cited of the transmission of the Avesta. There is a 
mass of other material also, much of it summarised, and therefore 
lacking apparent order or logical continuity. 

The Vlth book stands apart, being an anthology of wise sayings 
attributed to the Ancients and to various named Sasaman sages. Thanks 
to the clear andarz style, it makes the easiest reading of any (see further 
below, p. 51-f .). The Vllth book, which embodies many Avestan allusions 
contains a sort of universal history, from the First Man to Judgment 
Day, with Zoroaster as its central point.* The VHIth gives the famous 
summary of ig* of the Avesta "for the knowledge of the many" 
(0 agafok i wasan) ■ (one of the nasks was then already lost, and of another 
only the Avestan text remained). In the IXth book three of the religious 
[gatiamg] nasks are dealt with in greater detail. 

It is a striking fact that, except for Sdyist ne-sdyist, all the books 
so far considered are known, in their oth-century form, to be the work of 
priests and high-priests of Pars. Together they constitute an important 
part of the surviving Phi. literature ; and their attribution to one province 
indicates impressively the amount of such writings which must have 
disappeared; for, although Pars appears to have been the Zoroastrian 
stronghold of the qth century, it is hardly to be thought that there was 
not literary activity also in other provinces, whose products must later 
have been lost, through poverty and persecution. 

There is an anonymous work, probably belonging to the early decades 
0f the I0th centur y« which is generally transmitted with the Dddistan i 

- See de Menasce op. cit. 20-31. ' ' 

^ -rnukttd by ... C. C»s™,.u U If**, V „ 8S 6) ; ,„ 1>s „ ibcd by Da „^ v z „ pMm 

... 2&2S ■zsxzzsg&zssz s^isr * ,r ^ ssfa,; - e Dk - 

4 6 





denig, and is for this reason known as the Pahlavi Rivdyat accompanying 
the Dddistdnt, denig. 1 This is a text written 'in simple, generally correct 
Middle Persian, but later than the works already : considered (its unknown 
author cites, among other works, Sdyist ne-Sdyisl, Dddistdn I dmig, 
Bundahisn and Dinkwrd). It is an anthology of the Zoroastrian religion, 
treating predominantly of ceremonial, but also of ethics, customs, 
eschatology, and cosmology, and including fragments of epic, folklore 
and wisdom- literature. The sources used appear to he old and good. The 
emphasis on ritual probably marks the need to stress such matters under 
the pressure of Islam. 2 About three-fourths .of the- text were later in- 
corporated in the Persian Rivdyat of Kama Asa oi Carnbay. 3 

A short Pahlavi Rivdyat survives, which embodies the answers to 
questions put'to : £med I Asawahistan, high-priest in 955 A.C., and 
probably a descendant in the third generation of the redactor of the 
Denkard.* The questions are in the main practical, concerning legal 
problems about property, inheritance and marriage, or religious points 
of ritual and purification. The text has interesting features of vocabulary, 
with old words and idioms intermingled with others more familiar from 
the later Persian Rivdyats than from the Pahlavi books. Style and syntax 
are degenerate, as is to be expected at this late dab:. 

A o,th-century work which stands apart from those hitherto considered 
is the Skand-Gumdnig Wizdv, the "Doubt-destroying Exposition", 5 
composed by one Mardanfarrox i Ohrmazddadan, who, since he speaks of 

1 Ed. Dhabhab. The Pahlavi Rivayat accompanying the Dadisidn-i J'Hnik (Bombay, 1913), with 
English ititro. (and list of chapter-headings), q.v. for earlier work -in the text. Ch. I-LXII in 

Dkabhar's edition form the Rivayat preceding Dd., Ch. LXIII-LXV Ihe Rivdyat fcllo&itig Dd. — 
Since 1913 the following chapters have been transcribed and translated: I by Nvberg Texts sum 
mazd. Kalender, 44-47; XYIII by Kveerg in Oriental Studies in "honour of ... 5. C- E. Pavry 
(Oxford, 1934), 339-52; XLII by Tavadia in Modi Mem. Vol. (Bombay, 1930). 479-87; XLVI 
(except for 16-21) by Zaehner Zuruan, 360-67; XLVI II §§ 90-06,. ibid. ;i54-g; LXIV by de Mekasce 
Anthropos (1942-1945), 180-85, Almost the whole of the Riv. preceding Dd. was transcribed and 
translated, with notes, by H. K. Mirza as a thesis in London university in 19+2. 

2 See Dhabkar, intro., 2. 

3 See Dhabhar, intro. 3-3, and the introduction to his Saddar A r u?r and Saddar Bundehesk 
(Bombay, 1009), sis. In later centuries a number of Persian .Rivayat* were composed, also based 
on traditional material, and from these the Pahlavi "Rivdyats" havi: been named. [Most of the 
Persian Rivayats have been ed. by M. R. Unvala, in 2 vols., under '.he title Darab Hormazyar's 
Rivayat (Bombay, 1922), and translated by Dhabhar (Bombay, 1932:'.! 

4 See Wesi GIP ii 1:05-06; Dhabhar Ca,t. of the First Dastur Mchf'ji Ratio- Library, p, 15. The 
text was edited in 192S by B. T . Anklesaria, and in 1962 Vol. I of his work (tex t and transcription) 
was published in Bombay, as the Rimyat-i HimU-i ASavahtftdn. Vol. II (transl. and notes) is 
promised shortly. 

Another work with the character of a Rivdyat, but written in poor Middle Persian, is the Wizirkird 
i denig ("Religious Explanations"), which contains a number of Avustan quotations; see West 
GIP ii 69-90, Tavadia op. cit. 114-15. 

1 Ed. by de Menasce Skand-Gumdmh Vi6dr (Fribourg en Suisse, ^945). <!•"•> P- *+, for earlier 

his debt to the Denkard of Adurfambag, probably lived in the 2nd half 
of the century. In his introduction Mardanfarrox states that he had 
travelled far and wide to study different rehgions, and had come to hold 
as true the faith in which he was born. His book, designed to dispel 
doubt in others, falls into two parts: an exposition of Zoroastrian beliefs 
and a criticism of other religions (Islam, Judaeism, Christianity and 
Mamchaeism). What distinguishes it from the works already considered 
is that the argument is pursued on an abstract, intellectual level and 
most of the mythical and traditional elements which had attached them- 
selves to Zoroastnamsm are ignored. The style is elevated, but clear 
and balanced, the controversial parts .judicious, and temperate by the 
standards of theological debate. The work nevertheless presents difficul- 
ties, largely because of its transmission; for it survives only in a mediaeval 
Sanskrit version, and in Paauul i.e. Middle Persian transcribed in mediae- 
val times out of Pahlavi into Avestan script.* The corruptions during this 
process are sometimes considerable. 

^ There is a short work which also survives in Pazand, namelv the 
Urn i Kushg "Reasons for the Sacred Girdle",* which has some similari- 
ties with the Skand-Gumdnig Wizdv in both style and content Here too 
the treatment is largely abstract and philosophical; but the framework 
b traditional, in that a sage is represented as expounding the symbolism 
of the kush to a disciple, through question and answer. 

A few short texts directly concerned with xeligious life (such as praises 
blessings, prayers and confessional texts) survive in Middle Persian * 
but most such works have been transcribed,* or have been rendered into 
Persian or Gujarati, in which form they belong to the later history of 
Zoroastrian literature. 5 

Many of the texts still to be considered are also religious ones ; but the 
foregoing books form a group in that they belong to the period after the 
Arab conquest, either as derivations from the Zand or as original writings ■ 
and in that their authors share the common aims of expounding, conserv- 
ing or defending their ancient faith. 6 

tcxtllldTsS^^ — "J ■* P— 

S. D. Bhahvcua CU SankJ Writing %£ L^W^XtT" ^ ™ 

- Ed. by T. 15. Anklesaria in his Dandk-u Mainvii-i Khard fHrm.hai. ™r-,\. ,, a v. is •? 

jism^s^ss, ;sf TAVADiA op - cit - 12? ~ 8; h j ™ s & *■*** * 

! 5™ Iu* ?r 82a, |? V * CSi ° IlS S6C E - K - ANTJA Fdzettd '^IS (Bombay, rqoq). 

For the New-Persian. Zoroastrian, literature see West GIP ii 122-20 
• A IWd work which de lies classification is the Ao^dati*, which convolutions from the 

4 S 


4. Visionary and apocalyptic Texts 
Manticism plays a considerable part in the learning and literature of 
many ancient peoples, and the Middle Persian texts show this to have 
been true for the Iranians, as it was for their neighbours, the Indians and 
Greeks. One well-established type of mantic composition, "visions of 
the home of the dead," * is represented by the Arddy Wiraz Namag 
("Book of the righteous WiriLz"}.* This has proved the most popular 
of all Zoroastnan works, with verse and prose translations into Persian 
Sanskrit and Gujarati, and from them free renderings into European 
languages. 3 Its subject is the visions of heaven and hell seen by the priest 
Wiraz, after he has drugged himself in order to release his spirit and 
drsoover for his community the fate of the dead. He travels the path of 
the departed, crosses the dread Cinwat bridge, and is shown the joys 
of Paradise and the tortures of Hell. The latter are the more varied 
and form a grim and revolting array. Unswerving justice reigns in the 
hereafter, and every act is strictly recompensed. Having beheld these 
things, the spirit of Wiraz returns after seven days, and he relates his 
vision to those who have watched beside his body. 

The Arddy Wiraz Ndmag has probably a very old kernel (the name of 
the just WIraza occurs in the Avesta);* but when it was first set down 
is unknown. In its surviving form it is a prose work, written in simple, 
direct style; and an introductory chapter indicates a date after the Arab 
conquest. 5 This late redaction was evidently made in Pars, 8 and is 
probably one of the gth/ioth-century literary products of that province. 
Another branch of mantic literature concerns prophecy, "which may 

Avesta (possibly largely from the Hddozt Nask) on the theme of death, with MP. paraohiases and 

?rs:st r t t gip u > m ™" £R *■ *■ » »-*- -** ***** MStsvEi 

further studied by DucHESKE-GonxEMii.. J A 1036, avril-juin, .4,-55. T he text of 'he late Phi 
version was published by Dhabha* Essays on Iranian Subjects (Bombay, i 9 ^) 42 -6 2 The comply 
text wo edited, with transl. a*d notes, by KM. Jamasp-Asa for a doctoral thesis hx Bombay £ 

'« w S « r^t - K a Q T°^^ T ? G "! wtk ****»—" (Cambridge, ,934-1940) HI 848 II. 

* See W MT GI p u 8 . the edi has ^ appear<d o{ JaAs iArlviraf Nawk 

(Bombay, 1902}. I i-i 7 , s transcribed and translated by Baxley Zor. Problems 151-,, 
,1, k "T ™- aiatailied ^ at tbe AW - influenced the Mamk tradition of thekfraj. It has 
also been endlessly compared svith the Divin* Cmudy, The discovery of an Islamic source for 
Dante's inspiration (see G. Lew Della Vida * Attains XIV a (1940), 377 - 407l in Crea5CS the 
possibility of a link between the two. l !MW ' 37? 4 7) mcrea5CS tiie 

' ?" Q Bar tholohae 4ft«ucbf ir*tofa«*, 1454; the traditional reading of the name is 
JW, tat against Menace's argument in support of this (J A i 9w , 3-7) see Hnmnio *,««*, 
(Uxiord, 195 1 ), 51 n. 3. 

5 According to this, Wiraz is persuaded to enter his trance in order to establish the efficacy of 
various ceremonies (yaziSn, iron, afrlnagan etc.). This suggests a date when Zoroastrianisrn was 
TaJ^ ^ s "fk "*?«*■• A late date * confirmed by a suggestion of confusion between Wiraz 
and Wehsabuhr, high-pnest of Xosrau I.-On what seems to be a citation in the test from Aog 80 
see D ARMESrE T ER £l ir*n. II 75 f. . wu, enters j^ triaCe at tne Farnb S ft 


whom \ "*: ^;:f d Th tion is Jsmjsp ' z °~ ,s fot ~ 

Tta Av ,1^ " mparted th£ ««* of ^nation.* 

strong influence o n Greek tad f™ "or ^ **"*' eXerted a 

t; r^H£~ : ^ -^ar the Iron 
—^ more ^r^x.rrp^rr^: 

J Ckabwick op. cit. I 473 j see dso ibfaL m g 

binular prophecies are to- be found in oihpr n-»i u*™. 
see Chadw,™ op. cit. II 59 o-r; Rm^^ ^ 1 ' eratuws ' ****« Indian, Irish and Norse; 

sala, i 945 ) 4 6. 47> 49; 6o . ' A * Cf " rf Foia "•«* *«^ «« ^^(UA 0/ Co.? (Upp- 

* Zand i Wthman YaSt III 6-7- on thu rv 
mantic traditions, scattered over 'the world ^0^^"^^ ° P> dt> 45 " Ia a »™*« «>f 
source (see Cha»w ICK op. cit. I 640-50) 6n*S -. ^ ******* witter from a holy 
mythology s eeH . Lt, DERS ^«ii^l ^13^ To ^ 2 ^f ^r r^ ° f TrUth " ta ^-Iranian 
Mf* (Cambridge, x 959 ,, 7 ._ 0n ZoroasS'vat S^'IS?*? 1 ™ T»AmU» Hy^ u> 

Dk. IX 8 {West 5££ XXXVII, i3i} 
See Reitzenstein od. cit «fi it ■ w \jr, 
^Vni3( I9 ^ ;F . CuM P ONr ^y - H. * •™^J»^lj. Koninkl. Ak . u Amstefdam 

Bombay, : 9 o 3 ); and by B. T. A.«S 2^-5^ *?""" ^ (with ^^^ansl.) 
Wlh £n ^ sh *«-■). The printed copies o thf li t, StT^** " - ^ PaM ™ »^-* 
by fire m ro 45 , bat the edition was published in !«?? m* TS W I919 ' Were mostl y ^strgyed 
^ J. M. Unvala. Saoeq H»A«T^^u3ed 2ft ^"t a T ' AKKLESARIA = ^^ foreword 
transl. with notes in Tehran in i 94+ ed m1h B " J- A«*u*a*ia, published a Persian 

Handbuch der Oriental^, Ah, ,," Ed . I¥ , Abschn . 2 _ l% ^ C — ™ ^ W^« 93- 



pretation, given by Ohrmazd himself, of a dream dreamt by Zoroaster 
while he possessed the wisdom of all-knowledge. The longest section of 
the Zand i Wahman YaH is devoted to the Age of Iron, told with bitter 
feeling; this account is evidently partly traditional, and partly a lament 
for contemporary wrongs. 1 The text ends with a prophecy of the coming 
of Wahram i Warcawand, God of Victory, to restore Iran and the Good 
Religion, and to usher in the last rnillenram. 

The prophecy of an age of misery and wickedness, to be followed by 
the apocalypse, is found also in the Jamdsp Namag. 2 where it is foretold 
to King Wistasp by Jamasp the seer. 3 This text, fragmentary and badly- 
preserved, is in verse, 4 and thus in form belongs to an older tradition 
than the prose exegesis of the Zand i Wahman YaH. In content also it 
differs in details from this work. 5 There is no reason to doubt that the 
Iranian prophetic tradition was embodied, with minor variations, in a 
number of works, of which these two chance to survive. Its imprint is to 
be found also in the religious compilations. 6 

The Jamdsp Ndmag came to be incorporated in a longer prose work, 
the Ayddgdr i JdmdspZg, 7 of which it forms the r.6th chapter. Apart 
from the verse Jamdsp Ndmag, and one other Pahlavi fragment, the 
Ayddgdr survives only in Paz-and and Persian. 

The coming of Wahram I Warcawand is celebrated separately in a 
short poem, evidently of post-Sasanian date. 8 It has been suggested that 
prophecies concerning the God of Victory became blended in the late 
Sasanian period with tales of the heroic Wahram Cobin, to the embel- 
lishment, of the legend of the god.* 

1 On the aames of peoples in IV 58, VI 3,6 (A.'s edition) see Ba.ii.ev BSOS VI 4 (1933), 945 ff., 
XI 1 (*£+$), 1-4; Asiatica, Festschrift F. Weller (Leipzig, 1954), i3-*4- 

2 Ed. by Modi in Jdmdspi, Pahlavi, Pazani and Persian Texts (Bombay, 1903) and in part by 
West in Avesta, Pahlavi and Ancient Persian Studies in honour . . . of P.B. Sanjana (Leipzig, 1904), 
97-116; transcribed and translated by Bailey BSOS VI (1930-31), 55-85 and 581-600, and by 
Benvenjste RHR (1932), 337-80, 

3 On. the part played by WUtasp in. the Graeco- Iranian, apocalyptic tradition see BewvsNisrE 
op. tit. 377-78- 

4 This important fact, obscured by the Pahlavi script, and by the ancient conventions of Middle 
Persian verse, was discovered by Benveniste. 

5 See Benveniste op. cit. 368 ff. 

■ Notably GBd. XXX1II-IV, ed. Messina (see above), 

3 Ed. Modi op. cit.; Messina Libra afiocalitto -persiano, Ay&tkdr 1 2'ainasptk (Rome, 1939). 

8 This text is preserved in a famous codex, MK, which contains a number of miscellaneous works, 
some with only a slender connection, or none at all, with religion- The larger part of the codex is 
ed. by Jamasp-Asaha The Pahlavi Texts contained in the codex MK, II (Bombay, 1913), in whose 
edition the poem occupies pp. 160-1. It has been transcribed and translated by Baiiey Zor. 
Problems 195-6, and Tavadia JRAS r955, 29-36. Tavadia noticed that the text was a rhyming 
poem. A rendering of the text into Persian verse had earlier been made by M. Bahah, see his 
Divan I (Tehran, A. H. 1335), p. 548. 

' See K. Czegiedy Acta Orientalia Ac. Scieni. Hungaricae VIII i {1938), 20 ff. 



Spells, which play a large part in. the mantic literature of some peoples, 
appear only incidentally in the Pahlavi texts, and then in garbled form; 1 
but spells, divination, and the interpreting of dreams are prominent 
in the literature deriving from Middle Persian works. 2 


Middle Persian is rich in a wisdom-literature based on coUections of 
gnomes (brief, apothegmatic utterances relating to universal*) . This 
again is a widespread category of oral literature; and the three classes 
of gnomes which have been recognized 3 are all represented in the Pahlavi 
texts. These are 1) the gnome of observation ("there is no cure for 
age"); 4 2) the gnome of prudence or advisability ("make a friend of 
that man who will be most useful to you"); 5 and 3) the moral gnome 
("the best protector is one's duty").* The last two types merge readily 
into precept, as one of these examples shows, and the Middle Persian 
name for this class of composition is in fact "precept" [andarz). Naturally 
the Zoroastrian wisdom-literature, transmitted by priests, contains 
a high proportion of moral gnomes; it also embodies a considerable 
amount of doctrine, conveyed through statement or exhortation. Several 
of the andarz-texts, have a preamble in fairly elaborate style; but the 
andarz themselves are characterised by brevity and lucidity. The unit 
is the sentence, and there is usually no attempt at continuity, although 
separate aniarz are sometimes grouped by content. Some have a poetic 
quality, and it may be that the older Iranian gnomic literature (as em- 
bodied, for example, in the Baris Nask) was in verse. So far, however, 
only one verse-text has been discovered in the Pahlavi andar'z, and this 
bears signs of being a post-Sasanian work. 7 

Wisdom {xrad) is constantly praised in the andarz texts; but this is 
not the mantic wisdom of prophecy and divination, but that of obser- 
vation and reflection. Gnomes are characteristically old men's utterances 
realistic, prudent, ripe with experience. Those which are not anonymous 
are therefore commonly attributed to sages, or to men of authority 
such as kings and counsellors; and are often cast in the form of answers 
to question s put by some seeker after knowledge, a disciple or son. One 

, ' E.g. AZ. 41, 74. 

- Notably in the Sahname and Vis -u Ramin. 
' See Chadwick Growth of Literature I 375. 
* Jamasp-Asana Pahl. Texts 40. 15-16. 
1 Ibid- 59- 4-5- 
? Ibid. 39. 2-3. 

' Jamas?-Asana Pahl. Texts, 54; see Henning BSOAS XI1I 3 {1050), 641 ff. 





of the Middle Persian collections is attributed to Jam ( Yima), 1 another to 
Osnar the Wise, a Kayanian sage. 2 Most are assigned, however, to the 
Sasanian period. One collection is attributed to Xosrau I, s the others to 

priests or ministers. Prominent among these are Adurbad i Mahraspandan, 
high-priest of Sabuhr II; 4 his son Zardust; 5 and Wuzurgmihr I Boxtagan, 
minister of Xosrau I. 6 The andarz attributed to the last-named have a 
preamble in unusually elaborate style, which may be of his own compos- 
ing. 7 (There is no reason to suppose that by the 6th century Middle 
Persian was not capable of complex rhythms in the hands of a sophis- 
ticated writer.) At least one Pahlavi collection of andarz, attributed 
to Adurfambag i Farroxzadan, 8 must be post-Saxanian. 

The gnomic style creates a uniformity, and it is hardly surprising that 
"wandering" andarz appear in more than one collection. The various 
collections differ, however, in the proportion of th-dr components. Some 
are startlirigly full of gnomes of advisability, others are almost wholly 
religious and ethical. Among the latter is the great series of andarz 
which constitutes Book VI of the Dmkard. The first and longest set 
of these probably derives from the MP. translation of the Bans Nask; 
other sets are attributed to named Sasanian sages (among them Adurbad 
i Mahraspandan). There are 5 main groups, arranged according to 
the opening phrase of the andarz; and within each group there is an 
attempt at arrangement by subject-matter. The orderliness of 
this collection is unusual, and may owe something to the compilers of 
the original Avestan text. Many of the andarz, here and in general, are 
trite and dull; but there are sporadic touches of imagination, and moral 
utterances of high order. For example, the Zoroastrian emphasis on 

1 In Dk. III. — The relative clarity of tie andarz texts has led to their being much translated, 
and. it i$ only possible to give a, selection of references here. 

* Ed. Dhabhar Andarj-i Ao-shnar-i DS-nak (Bombay, 1913); on Oinar and the "Hosang" book 
of Islamic tradition see Heknihg ZDMG 1956, 75-6- 

1 This, like many andarz- texts, is in the codex MK; ed. Jakasp-Asana Pakl. Texts 55 ff.; trans! . 
{with 4 others) by J. C. TaRapore Pahlavi Andarz- Namak (Bombay. 1933). 

* More aw&rz-collections are attributed to Adurbad than to anyone else. There is one in Dk. Ill 
and others in Dk, VI (see Zaehher J JR. AS 1940, 36 ft, Zurvan 407-S}; one in the Pahl. Jiiv. Dd. 
LXII (Djsabhar, p. 193); and two in codex MK (Jamasp-Asana PzhLTexts 58-71, H4-53)- The 
first of these two is txansl, with notes, by Fr. MOU.ER Sb. KA W. Wwi CXXXVI (r897), 1-25, the 
second by S. K, Meherji Rana (Bombay, 1930). Both are transl. by Zaehner The Teachings of 
the Magi (London, 1956), 101-16. 

8 To him is ascribed the Pandnamag i ZarduSt, also called Cidag andarz i poryotkeidn, ed. Jamasp- 
Asana Pahl. Texts 41-50; A. Freiman WZKM XX (1906), 149-6", 237-80; Nyberg Hilfsbuch 
des Pthlevi, 17-30; and M. F. Kanga, with transl. and notes (Bombay, i960); transl. Zaehner 
op. cit. 20-28, and H. Corbin Poure Davoud Mem. Vol. II 129-60 (q.v. for further references). 

« Jamasp-Asana Pahl. Texts 85-101; see ChhiS*£nSEK Acta Ori-enialia VIII (1930), 81-128 
(with references also to Islamic renderings of the andarz of Wuzurgnuhr). 

' See Henning ZDMG (1956), 77- 

* Jamasp-Asana Pahl. Texts 79-80- 

ritual purity (with avoidance of dead matter) has often been stressed as a 
barren element in this great religion; but the following andarz sets in 
perspective these merely physical demands: "Keep further from doing 
injury and harm to any than you would from the corpses of men; for 
the pollution and filth which comes to the body can more easily be 
washed away and cleansed than that which comes to the soul" (az 
azar t(d bes I pad kasan duriar -pahrez ku az nasay % mardohmdn; ce dludagih 
ud rimanih i tan rased sustan ud pak kard&n xwdriay sayed ku an Id 
nwdn rased DkM. 581 3 - 5 ). Not even in the Dk. collection are all the 
andarz concerned with religion, however; and it is evident that a number 
of them are simply the surviving representatives of a popular type of 
oral composition, which have become associated with religious works, 
and have thus chanced to survive. In course of time the less religious 
andarz, with all Zoroastrian elements excised, were incorporated in the 
Islamic adab literature, and lived on in a new sophistication and elegance. 1 
Two general points have been made with regard to the andarz. One is 
that the fatalism characteristic of many stamps them as a product of the 
Zurvanite branch of the Zoroastrian church. 2 Fatalism is, however, 
typical of the genre. The sage, surveying realistically the human lot, 
sees small chance for man to be master of his own destiny. A fatalistic 
element is probably, therefore, part of the non-religious character of many 
of the andarz. The other point is that the constant counselling of prudence, 
the mean of action, marks a debt to Greece, and to Aristotelian philo- 
sophy. 3 Again, however, prudence is in general recommended in hortatory 
gnomes, not only in Persia. Man's best course in a precarious existence, 
so the sages say, is to steer carefully, avoiding those extremes likely to 
hasten disaster. The philosophic content of such gnomes is moreover 
slight. Except in specifically religious utterances, it is the results of action 
rather than its causes which interest the Iranian sage. It is, however, 
conceivable that in selecting gnomes for written record the Zoroastrian 
priests were influenced by knowledge of the philosophy of the mean, 
whose practical applications to conduct were thus illustrated in simple 

With religious gnomes, exhortation readily blends with exposition of 

1 See M. INOSTRAN2EV, transl. by G. K. Nariman, Iranian Influence on Moslem Literature 
(Bombay, 1918); Zaehner Zurvan 40S; Henning ZDMG (1956), 73 ft'.; Minovi Rev. de la Faculty 
<Us Lettres dtl'Univ. de Teheran IV 2 (1957), 58 if-; a doctoral thesis on the andarz literature, with 
special reference to this aspect, was presented in London by S. Shaked in 1964. 

3 The nature and influence of Zurvanism has been much discussed in recent years; but this 
problem belongs to the theological lather than to the literary field. 

3 See de Menasce Un-e Encychpidie mazcUenne 38 ff. 



doctrine, and the apothegmatic style then necessarily yields to a simple 
narrative one. It is thus possible to regard the DMist&n i Mendg % Xrad, 
"Judgments of the Spirit of Wisdom," i as an extension of the religious' 
undarz, with the embodiment also of a certain amount of antiquarian 
learning". In some respects this is a classic example of andarz literature. 
There is a preamble; the body of the work is in the form of question 
and answer; and many of the answers are expressed as gnomes. On the 
other hand, the questioner seeks (according to the- preamble) to establish 
the truth of Zoroastrianism; he receives instrucrion from the Spirit of 
Wisdom itself; and the text is unusually long, with a large doctrinal 
element. The work is, accordingly, far from typted in its general charac- 
ter. There are several small indications of its Sasanian date, among them 
a reference to the Zoroastrian inquisition, 2 which accords closely with a 
passage in the Letter ofTa-nsar, which belongs partly to the 6th century. 3 
The preamble too fits well with the age of Burzofs. No doubt these, and 
other, indications would suit some earlier Sasanian reigns as well; but 
on the assumption that the use of writing extended slowly, in the wake 
of the written Avesta, the Mendg i Xrad may be assigned with some 
confidence to the reign of Xosrau I.* It is perhaps an ecclesiastical 
counterpart to the political treatises of this period, which also appear 
to owe something to the andarz tradition. 5 

Another branch of wisdom-literature, general through the world, is 
the riddle. Middle Persian has a literary example in the Madiyan i Joist % 
Fry an* in which is related the contest between Axt, a sorcerer, and the 
Zoroastrian Joist, who must answer the riddles propounded by Axt 
or lose his life. He resolves them all; and Axt, failing to answer his coun- 
ter-riddles, dies instead. There is no internal evidence of date; but both 
Joist and Axt are mentioned in the Avesta, so that whatever the period 

. A ~t "~° ?; to the works thexe atoned may be added the edition of the Pahlavi 

text by P. Sanjana The Dhid i Mainu i Kktal (Bombay, 1895); and of the Pahlavi, Pazand and 
Sanskrit texts, by T. D, Anklesaria Ddndk-u, Mainyb-i Kkard (Bom-bay, 1513} The Pazand 
version is given by An*ia Pteend Text, 273.33.,. On the title see Bailey Zor. Problems % n 3 A 
number of indi vidua: passages have been edited separately, notably by Nvberg J A x 9 sg i 108 ff 
and ia HUfsb-uth <Us PehUvi; Zaehner Zuruan (see index, p. 4 6r},— The Minbg i Xrad was 
apparently preserved in India, whence a copy seems to have been brought to Persia, perhaps ia 
the nth century. In India itself its Pazand and Sanskrit versions .mly have survived, together 
with later retranslafdons into Pahlavi of the Pazand text. 
' MX. XV 22-25. 

s Ed, M, Mznovi (Tehran. 1936) P- 17.1 ff.; see Daemesteter ZA I 226 n. 2, and see further 

« This was West's tentative opinion (see his edition, x-xi), based en such passages as MX. I 18 
(a reference to Mazdak?) and the whole of XV (on kingly power). 
5 See below, p. pp. 60-61. 

• See West GIP ii ro8. 



of the written Middle Persian redaction, the kernel of the work is likely 
to be old. 

Riddle-texts have the element of contest. Another "contest" work 
(also belonging to a well-known oral category) is the Draxt i asurig 
("The Babylonian Tree'V This text is a Middle Persian redaction of a 
Parthian original, 2 and is in verse. 8 It concerns a contest over precedence 
between a date-palm and a goat; and is full of difficulties, due to the- 
dialect-mixture, verse-form, unusual vocabulary and a riddling element. 
As well as being a contest-work, this is also incidentally a catalogue-poem, 
listing the qualities of tree and animal. It belongs, therefore, to wisdom- 
literature, being intended both to sharpen the wits and to give instruction. 

6. Non-didactic Poetry 

The works so far considered were evidently transmitted by priests, 
the scholars and teachers of their age. Non-didactic poetry was created 
by the minstrel, who played an important part in Sasanian society: 
"entertainer of king and commoner . . . present at the graveside and at 
the feast; eulogist, satirist, story-teller, musician; recorder of past achie- 
vements and commentator of his own times." * The minstrel-poetry was 
sung, usually (if not invariably) to an instrument; and minstrels of the 
first rank, such as the great Barbad (court-minstrel of Xosrau II) were 
composers of both words and music, and by the beauty of their minstrelsy 
exerted great influence. In late Sasanian times the king's minstrel was 
numbered among the four chief men surrounding the throne. Poetry for 
entertainment was also cultivated by women and boys of rank, and 
poetry was used spontaneously and generally to express emotion. 

It was possibly in part its close association with music which kept 
poetry unwritten until well after the Sasanian period; and, the music 
lost, it may never be possible wholly to appreciate Middle Persian ver- 
sification. A key to its principles, however, has been provided in this 
century by the Manichean hymns.* In these the verse is governed by 

' Ed - J*™-A8a*a fjrtt Texts 109-r* and Unvaia SSOS II iv (1923) 637-78 (q.v, p. 6,8, for 

references on "Rangstreif'-Literature). " ,F 

» See Barthoiomas At Kenntnm d. miiUUranischen. MvndarUn IV (1922) a 3 f. The Parthian 
original must have oeeu unwritten, and the Middle Persian rendering probably also remained in 

^S^THi i. I!" the AxA C ° aqUeSt ' Mixtures of dialect-forrns such as the text shows 
can be paralleled m the borrowings of texts between related languages in known oral literatures. 

™J5?JT? rTv? W* WV * HI f E ' See JA *«° ^ 193 ff-; the text was the first poem to be 
"cognized in the Pahlavi MSS. On the versification see further Hen king BSO A S XIII 3 (1950) 
* xdV? SOtne adfhtJonal notes oa tie text see G. Bolognesi RSO XXVIII (3953) 174 it 
JRAS 1957, 18 (q.v, pp. IO -45, for a detailed study of Iranian minstrelsy) 

in SrJ MW -, Py - *"* L ° nd0n (I94?) ' 5 *"5 6 ; Bovc* The Manickaean Hymn-CycUs 
*» Parthian- (Oxford, 1954), 4.5-59. 



stress, not quantity; and in each hymn the number of unstressed syl- 
lables varies, within limits, from line to line. There is a variety of metres 
(whose rules are not yet fully understood); but rhyme is lacking-. The 
metrical principles deduced from these hymns apply also to the poems 
in the Pahlavi books, 1 from which it appears that Sasanian poetry in 
general was composed in this convention, which is one of high antiquity. 
After the Arab conquest this versification gradually yielded to a new one, 
based on Arabic models— rhymed and quantitaiive, and written down 
by the composer. As this slowly gained ground, the minstrel-poems, 
still largely unrecorded, were either refashioned, or fell into neglect' 
Only a few short pieces survive in the oldest Persian books, 2 and even 
these are influenced by the new convention of rhyme. The process was 
very slow, however, and minstrel-poems were still sung in the nth 

Only one category of Sasanian poetry, the narrative poem, survives 
both in its own right and as an element to be traced in the later literature. 
This is largely because one epic cycle, celebrating the Kayanians, was 
linked with religious history. Its stories appear to have been little known 
in Pars at the beginning of the Sasanian era, and the Persian minstrels 
evidently acquired them from Parthian singers, in the north-east. There 
they had become interwoven with tales of Parthian warriors and Saka 
heroes, 3 a process which probably took place unconsciously towards 
the end of a long period of oral transmission.* Of t his cycle only one frag- 
ment survives in its original verse-form, namely the Ayadgar i Zarerdn 
("Memorial of Zarer"). 5 This tells of the battle for the faith between 
Wistasp and the pagan •Xyons, in which his brother Zarer perished. 
The fragment was evidently written down (presumably after the Arab 
conquest) 6 because of its religious interest; but apart from stereotyped 
allusions to the Mazdayasrian faith, it is entirely secular in spirit, and 
in style is characterised by the fixed epithets, hyperboles and repetitions 
of oral epic. It has clearly lost in being written down, and has suffered 

1 See Kenning BSOAS XIII i {1950), 641-S. 

2 See U. M. Daudpota in. /. /. Modi Mem. Vol. (Bombay, 1930), ;4i-si; Henning in Handbook 
I, IV i 87 n. r! 

3 See Noldeke Das iranische Nalionalepos, lite Auflage (Leipzig, 1920), 7-9; J. Markwart 
Cauczsica VIII [1931), 78-113. 

* See Serta Canldbrigiensa, F. Steiner Verlag (Mains, 1954), 49-51; BS04.S XVII 3 (1955), 473-4.. 
s Ed. Jamasp-Asana Pahl. Texts, 1-16; transcribed and translated by A. Pagi-taro 11 iesto 

■pahlavico Ayatk4r-i-2artran (Rome, 1935}. 

* As Nolbeke has pointed out (op. cit., 6} the wax against the Xyoas, which in Tatari and the 
Sahndtne is long-drawn-out, is foreshortened in the Ayadgar to a single episode. This abbreviation 
probably took place in oral transmission, continued for centuries after the compilation of the 
Bock of Kings. 



corruptions through careless copying ; but in it can be heard the authentic 
ring of the old poetry. Similarities in style show how much Firdausi owed 
to the old epic tradition, still living, without doubt, in his day 

Another narrative poem. Wis u Ramin. is also recast, as it happens, 
from a Parthian original.* This recounts, at length and with a wealth of 
warlike and romantic detail, the love of the two persons from whom it 
is named. The mmstrel-poem was written down in Pahlavi script some 
time after the Arab conquest, in a version compiled bv "6 wise men" 
The poem evidently continued in oral currency for generations after- 
wards; and the written version was still used for the studv of Pahlavi 
in the nth century. 'It then attracted the attention of Gurgani, who made 
of it a new poem.* A Georgian Version" also exists. StylisticaUv the min- 
strel-poem appears to have had much in common with the epic; but the 
material belongs to a later age, probably the first century of the Christian 

7. Historical writing 

Middle Persian works are necessarily largely traditional. An oral lit- 
erature is static compared with a written one, the scope for innovation 
being small when each generation has to memorise as well as to create 
As the use of writing spread, however, the Sasanians embarked on new 
ventures, notably in the field of history. Traditional learning furnished 
many materials for writing history. There existed evidently genealogies 
bnef chronicles, legends about the origin of different Iranian peoples' 
catalogues of battles and great disasters, lists of place-names and river- 
names (with speculation about their origin and local events attaching to 
them), traditions concerning festivals and customs-in short a great 
mass of antiquarian learning, some of which is preserved in the surviving 
Zoroastnan books. In the prophetic works, moreover, was embodied al- 
ready a stylised history of the world. For generations, evidently, the Ira- 
nians had sought to impose patterns upon the confusion of events and the 
multiplicity of world-phenomena. With the development of writing they 
obtained the tool necessary for a more elaborate synthesis. The result 
was the composition, during the late Sasanian period, of. the famous 
Xwaday Namag Book of Kings". This great chronicle survives mainly in 

^£Zr5to5°£* (I94&) ' 7 ^ 4 < XI1 (I ^< »* ™ <■*«), 9«; h™c 





Arabic derivatives, and, less directly, in the Sdhndme. 1 It is impossible 
from these to judge it stylistically, but together they probably give an 
excellent idea of its content. Naturally in these Muslim redactions the 
Zoroastrian element is reduced to a minimum ; but it can hardly be doubt- 
ed that the original was the work of priests. 2 In it the reigning dynasty is 
glorified, and the history of Iran set out through the succession of her 
kings; but dynasties and events are shaped on a Zoroastrian pattern. 
The focal point is the life of the prophet. To honour the first champion of 
the faith, Vvistaspa, his dynasty, the Kayanian, had been linked by des- 
cent to that of the Pisdadian, one artificially composed of gods, who were 
represented as culture-heroes and early kings (the linking of these 
"dynasties" was pre-Sasanian, and provided the framework for the 
Avestan interpretation of world-history). Now the fame of the Kayanians 
was used in turn to enhance that of the house of Sasan, their supposed 
descendants. The truer forerunners (geographically) of the Sasanians, the 
Achaemenians, exist in the chronicle only to provide, with the first and 
last Darius, an artificial link between these two dynasties. 3 The chrono- 
logical framework also is Zoroastrian; and it was the exigencies of the 
Zoroastrian "world-year" which obliged the compilers to reduce by almost 
half the epoch of the Arsacids, 1 of whom in any case they had but scanty 
knowledge (based on confusing king-lists and brief oral chronicles). 5 
Evidently ^as a historian, no less than as a scientist, the priestly scholar 
was sometimes trammelled by dogma. 

The Arsacids play nevertheless a vigorous, if unacknowledged, part 
in the chronicle; for the priests, to enrich the bare sequence of royal 
successions, evidently drew for the early period on [he Kayanian minstrel 

1 FirdausL's Sakndme (and also Tha c alibi.'s Gkurar atybar muluk al-furs) derives from & work 
which seems to have been, no more than, based on the chronicle (see NIoldeke, op. cit., 17 ff.). For 
the older preface to the Sd/mdme (with references to some of the literature since Noldeke} see 
Miuoksky in Studi OrierdalisUci in (more di G. Levi deila Vida- (Rome, 1956)1 s 59-79- 

* Through the Xwaddy Na-mag a Persian national tradition was created; but to call it the 
representative of such a tradition is misleading, although clearly there is a valid distinction to be 
drawn between it, a composite work fashioned partly for secular reasons, and the purely ecclesiastic 
tradition of the Avesta and Zand, See particularly Christensen in Les Kayanides and Les gestss 
des rois dans Us traditions de I* Iran antique, Conferences Ratanbai Kelrak (Paris, 1936}. 

3 Artaxerxes Longomanus can hardly be said to figure in the epic, since his individuality is 
submerged in fiat of Wahman, grandson of Wistispa. Knowledge of him appears to have been 
derived from a Syriac source (see Noldeke op. cit., 13). It is not to be supposed, however, that 
there was any deliberate suppression of Achaemenian material by the priests. Probably no chron- 
icles or king-lists survived from that period, to which to attach such traditions as were still extant. 
Had organised information existed, the priestly compilers would undoubtedly have felt Obliged to 
accoramodate it somehow in their scheme of history. Their conscientiousness in this respect is 
shown by their inclusion, however summary, of the Arsacid dynasty. 

* See H. Lewy JAOS LX1V (1944), 197-214; Tacuadeh JRAS (19^7), 33-40. 
6 See H. Lswy, op. cit. 

epic. 1 They had, however, no criteria of criticism by which to disentangle 
the stories in these minstrel-poems; and so the Saka Rustam and for- 
gotten Arsacid princes — Frahad, Godarz and the rest — entered the chron- 
icle in Kayanian company. 

The compilers of the Xwaddy Namag used their epic sources well, 
and the fire and imagination in these make the Kayanian period the 
most vivid and remarkable in the chronicle. They drew also on wisdom- 
literature, providing each king with an enthronement-speech in the 
andarz tradition, and thus imparting, even to the era of barbarism, a 
moral element. The strain of resignation already noted in the andarz 
fuses readily with the epic feeling for the evanescence of worldly glory; 
and there may also be an element of Zurvanite fatalism in the compound, 2 
Doubtless in the Xwaddy Ndmag itself the certainties of the hereafter 
were stressed as a corollary to earthly vicissitudes; but in the Muslim 
redactions such religious consolations are little emphasized. 

The compilers further used written foreign sources where such were 
available, notably a Syriac version of the Alexander-Romance. 3 The 
account which this gave of Alexander was at odds with Zoroastrian 
convictions, but it was nevertheless incorporated, and the discrepancies 
allowed to stand. Heirs to an age-old oral tradition, with its tendency to 
conserve rather than to question, the authors of the chronicle had neither 
the tools nor the training for a critical examination of sources. They 
aimed, not at testing and rejecting, but at giving coherence to a mass of 
material connected with a chosen sequence of events. Everything which 
seemed relevant was included. Their intellectual achievement lies in the 
assembling of material; in the order and clarity they imposed on heter- 
ogeneous — and often basically unrelated— matter; and {fox the events of 
their own epoch) in a growing feeling for historical narrative. They not 
only provided the Sasanian kings with a long ancestry, but Persia with a 
history on the grand scale. On this foundation Muslim historiography 
was largely based. 

In the prose work, compounded from Pahlavi materials, from which 
the Sdhndme derives, two pre-Sasanian legends of genuinely Persian 
character were evidently incorporated. One is the story of Zariadres 
and Odatis, included apparently because of a confusion between Zaria- 

1 This appeare to have taken place at about the mid- 5th century, see Noldeke op, cit., 5; on 
the use of the epic material see further the references given above, p. 55, n. 4. 

' See H, Rikooren Fatalism in Persian Epics (Uppsala, 1952). There is at least one explicitly 
Zurvanite passage remaining in the Sdhndme, see I. P. Blub in Indo-Iranian Studies honour 
of ... P. Sawjana (London, ra^h 61-2; Zaekker Zurvan 444-46. 

s See Noldeke op. cit., 12. 



dres and the Kayanian Zarer; 1 the- other a romantic account of Ardasir I, 
which replaces the more factual version of the official chronicle. For- 
tunately the Pahlavi original of this survives, under the title Kdmdmag 
I Ardasir ("Book of the Deeds of Ardasir"). 2 This is a short prose work, 
simple in style, probably written in Pars towards the end of the Sasanian 
period. 3 It too was evidently the work of priests, 4 and a comparison of it 
with Firdausi's rendering shows how effectively Zoroastrian elements 
were obliterated in the Muslim-redaction. The Ka.rm.mag contains some 
historical details; but its generally romantic character has been explained 
as due to contamination with legends of Cyrus th=5 Great, still current 
then in Pars. 5 

The Kdmdmag was not the only prose work devoted to what was an 
episode in the chronicle. Ibn an-Nadim lists, 6 for example, separate 
works on Xosrau I, AVahram Cobin, 7 Xosrau II, , and Rustam and Is- 
fandiyar. These evidently also all belong to the later Sasanian period; 
but whether they were sources for the Xwaddy Ndmxg, or derived from it, 
or were wholly independent, there is little means of knowing. The Xwaddy 
Ndmag itself brought the history of the Sasanians down to the death of 
Yazdegird III. After the Arab conquest a number of separate works 
were written, dealing with the war and local campaigns. 8 

8. Political Treatises 

From the court of Xosrau I emanated two political treatises, which 
survive only in Arabic and Persian translations. These are the "Tes- 
tament of Ardasir" 8 and the "Letter of Tansar." 10 As their names show, 

1 See BSOAS XVII 3 (1955), 463 «• 

2 This is preserved in the codex MK. It has been ed. and transi. by D P. San j ana (Bombay, 
1896}, K. A. Nosherwan (Bombay, 1896} and E. K. Antia (Bombay, laoo); and was earlier transi., 
with notes, by Noldeke Bc-zzenbergers Beitrage IV (1878), 22-69; see also Fr. MCller Sb. KAW 
Wien, CXXXVI (1897), VI. Ch. I-III have been re-edited by Nyberg in his Hilfsbuch des Pehlevi; 
and the text has been published in Persian transcription, with note; (together with the Zand i 
Wahman YaSt) by Sadeo, H£dayat, Tehran 1944. 

5 See Noldeke op. cit., 23 ff. ; it seems likely, however, that it survives in a later redaction, 
since the language is a very late Middle Persian. 
* Ibid., 27-8. 

s A. v. Gutsckmid ZDMG XXXIV (1880), 585-7 = Klein* SchrifUn III 133 ff. 
' Fihrist, ed. G. FltDgel, 305.4 ff.; see Christensen Les gestes des rois 57 11. 

3 See Christensen Romanes om Bahrain Tchobin (Copenhagen, 1907), 
8 Fihrist, loc. cit. 

' Selections from this treatise survive in. a number of Arabic works, and in the Sakn&me (see 
Christensen Les gestes des rois 90 ff.). The longest extant version is- t-iat given by Ibn- Miskawaih 
in his Tajarib al-Umam, published in facsimile in the Gibb Mem. Series Vol. VII, 1 (1909) pp. 99 ff., 
and printed by Dehkhoda Amsal va Hikam III, 1615 ff. 

10 This survives in Ibn Isfandiyar's 13th-century Persian translation from the Arabic of Ibn 
Muqaffa 1 . It has been ed. by Darmestetee J A 1894, 185-250, 502-55; by Mjnovi, (Tehran, 1932) ; 



both were attributed, for greater prestige, to the first reign of the dynasty, 
and they evidently contain a core of matter transmitted from that 
period. 1 Both are whole-heartedly Machiavellian. All things are justified 
in the interests of authority and order. Church and throne are twins, 
linked in aims and activity, and subjects should suhmit to their bene- 
volent yoke. Passages which justify bloodshed, or the king's caprice 
(exercised to prevent in his subjects too dangerous a self-assurance) 
make chill reading; but both works contain much of interest for Sasanian 
society. They resemble one another closely; but the "Letter of Tansar" 
has been made more palatable by the insertion of two Indian fables, 2 
The Persian and Arabic translations are clearly much transformed in 
style, so that it is impossible to know the character of the originals. In 
both, however, the anda-rz tradition is in evidence. 

9. Law-books 
It was a Sasanian tenet that even as church and crown were joined 
together, so religion and the law were indissolubly linked. The judges and 
lawyers of Sasanian Persia were priests, and for tins reason many of the 
surviving seals (for affixing to documents) are priestly ones, "in late 
Sasanian times there must have existed numerous works on law, for 
even in the gth century Manuscihr writes of consulting "many law- 
books" {was dadistan-ndmag).* Only one survives, however, and that in a 
single, mutilated MS. This is the Mddaydn i Hazdr Dddisidn ("Book of 
a Thousand Judgments").* This lengthy work appears to have been 

and by A. I G bal in hise dition of the Ta'rix-i TabarisUn (Tehran, 1942). On the 6th-cent OT y dating 
see Christensen UEmpire <U $ Sa-ssanicUs (Copenhagen, 1907), m-12; Mikovi, preface to bis 
edition. j 

<■ See the introduction to an English translation of the Letter of Tansar by the present writer, 
home Oriental Series, Literary and Historical Texts from Iran I, 1967. 

* See Asia Major n,s. V i (1955), 50 ff. 

* Dd. Pure. LVT, ed. Dhabhar The Persian Rivayats, p. 184, transi. p. 188. On the Christian 
Persian, law-books see E. Sachau Syrische RecktsMicher III (Berlin 1934) 

* Tht Lv^ 6 *?• Wa -, a ! : S0Die time SpUt int ° lwo - The first 55 folios (which were disarranged) 
were published in facsimile by J. J. Modi MUigan-i Hazdr Dddisidn (Bombay, :90s). The remaning 
folios, numbered 74-91 (in proper order) were published in facsimile by T. D. Ankles aria (with an 
S*"™ y t. ° DI) Ti ? I ™ 1 COde ° J tht FafSiiS in SasllniAn T *™s or the Madigan-i-Bwr DMistar. 
II (Bombay, 1912). There is a translation by S. J. Bulsara The Laws of trie A»c*nt Persians 
(Bombay, 1937). The work has been extensively studied and commented on by Bartholomae, 
notably in 5 papers "Zum sasanidischen RcchP • I-V in Sb. HeiMberger AW., 1916-1923 {q.v., 
1918, p. 3 for references to earlier papers); Die Frau im sasanidischen Reeki (Heidelberg, 1924). 
Bartholomae maintained that the work defied complete and adequate translation. For other 
papers on this text see A. Pagliaro XSO XV (1935). 27s -315; XXIII (1948), 52-68; N. Pigulsvs- 
kaya m Papers presented by the Soviet Delegatus at the XXIII International Cmgress of Orientalists 
1954 (Section Iranian, Armenian and Central Asian Studies); J. DE Mbkasce in Indo-Iranica 
Merges f resettles a G. Morgensticrne, (Wiesbaden, 1964), 149-54, Dr. J. M. Un-vala Medial 
voluw (Bombay, 1964), 6-n and Feux el fondaUom pieuses dans U droit sassanide (Paris, 1964); 



compiled by one Farroxmard I Wahram, possibly in the reign of Xosrau 
II, the last Sasanian sovereign named in it. It is wholly Sasanian in char- 
acter, and in it are cited other law-books and many individual commen- 
tators (including authorities quoted in the Zand). 1 As its title implies, 
the work embodies case-history (actual or hypothetical) rather than a 
systematic legal code. The judgments given are in the main concerned 
with civil laws, affecting marriage, inheritance, property, rents, trade 
etc., which could still be observed by the Zoroastrian community under 
Islam. The technical nature of its contents, the state of the MS., and the 
usual careless copying make the work very difficult to interpret. 

A specimen of a marriage-contract {fiaymdn s zan grifian), dated A. 
Y. 627 {=1278 A.C.), is preserved in the codex MK. 2 

10. Short didactic Prose- works 

A number of short miscellaneous treatises survive, mostly in the 
codex MK. Some are traditional, others deal with r.ontemporary matters. 
One, a representative of the old oral category of place-name catalogues 
(cf. GBd. 80 1 , ayddgdrihd I §ahrihd), is the Sahristunihd I Erdn, z in which 
are enumerated the chief cities of Iran and the legends attaching to 
them. This treatise was probably first written down in the 9th century, 
after the time of the Caliph al-Mansur; but it contains good old tra- 
ditions, with the usual mingling of predominant!:/ Kayanian and Sasa- 
nian elements. The work reflects the interpretation of history given in the 
Xwaddy Ndmag, but with the small modifications of its longer oral 

A work of somewhat similar character, the short Abdih ud sahigih I 
Sagistan ("Wonders and remarkable features of Soistan)" 5 recounts the 

Boyce, BSOAS XXXI i and ii {1968). An edition of the -whole text ;s in preparation by A. Peri- 
khanian in Leningrad, 

I See Darmesteter in the iatro. to Anklesaria's edition, pp. 10-11. Darmesteter and West 
■both saw in a certain Juwanjam, cited several times, the oth-<£n;.ury father of Manusclhr and 
Zadspram; but nothing else appears to beer out such a late dating for the work. 

* Jamasp-Asana Pahl. Texts 141*3, transl. ibid. 47-9; see Tavajua Die m.-p. Sprtxchs u. Lit., 
108-09, Transcribed with Russian translation and notes by A. Pebitcharian in Sovets/taja Eltw- 
grafija,, i960, no, 5, 67-75; see further A. Perikhanian and D. N. Mackenzie in the K. R. Carta 
Oriental Institute Golden Jubilee Mem. Volume (Bombay, in the press). 

» Jamasp-Asana PaM. Texts 18-24; text, trawl, and oomraeniary by Ma&kwart pub. by 
Messina A Catalogue of the Provincial Capitals of Branshahr (Rome, 1931), q.v., p. 5, for other 
publications. On §§a-5 see further S. Kiva in. MajalU-yi Dani&hi&t-yi Adabiyyai, Tehran, II 3 

(1334/1955), 47-9. 

- See Markwart, op. cit., 6. - 

II jAMAsr-ASANA Pahl. Texts 25-61 transl. by Modi (with Ayadga-r i Zarcran and Sahrist&mhd i 
Bran), Bombay, 1899- West's transcription with translation was published posthumously by 
£. W. Jackson in J A OS 36, rgi6, 115-21. 



natural features and traditions which made Seistan pre-eminent. Among 
these is claimed the oral preservation of the Avesta after Alexander's 
conquest. The text is corrupt, and hears no indication of date; but it 
has the distinction of being the only Pahlavi text composed in a province 
which demonstrably is not Pars. 

Another catalogue- work is the Mdh i Frawardin roz i Horiadi in 
which are enumerated the great events which have taken place, or will 
take place, on the 6th day of the ist month, from the creation of Gayo- 
mard to the resurrection. This text appears to have been written in the 
reign of Xosrau II, although it certainly embodies much older priestly 

Two of the treatises are wholly Sasanian in matter. One, the Wizdrisn 
itatrang* describes how the game of chess was sent from India to 
Xosrau I, as a test for Iranian wits; and how the wise Wuzurgmihr 
both solved it and invented as a counter-challenge that of New-Ardaszr 
(nard). This game is interpreted as having an elaborate religious sym- 
bolism. The other treatise, Xosrau ud Redag ("Xosrau and the Page) " » 
is also set m the reign of Xosrau I, and may well have been actually 
written at that time. It tells how a boy of family presents himself, an 
orphan, to the king, describes his own lineage and training, and asks 
to be put to the test. The king sets him a series of questions, on what is 
best m food, wines, music, scents, flowers, women, horses etc. The 
page's answers create vividly a world of courtly luxury and the refine- 
ments 01 sensual delight. Nevertheless there is reason to regard this 
work also as of priestly redaction. In genre it belongs essentially to the 
wisdom-literature : there is the theme of trial by knowledge, and instruc- 
tion is conveyed in every answer. There are, moreover, some overt priestly 
touches. The boy's mother is of priestly family; he has learned the Avesta 
and Zand by heart; and when asked finally if he himself greatly desires 
material things, he answers that his sole wish is for iooo good deeds, such 

-, 5 J A! 2: AS t' A t SAXA P f L Tt **? I *' 08 '' transL by E ' B «™rin Revue arcktologiqw (Paris iSoO 
i?: « ; K. J. in K.R. Can*, Mem. Vol. (Bombay, xooo), X22 . 2 * ^ ' ' ™' 

^AZTiZ^fB^VlT ST;' P '- 40 ' fOT a "^ Wation,; transcribed and 

the source used by Firdauri and Tha'Jfbi ' ' 95l) ' ^'"^ ^ ttMiuB W3S emb «^ * 

Mnifkad+yiAtoUyya, Tabriz, VII ™^*™°'™** bhahei ™tem*mtheNa£riyy a M 

6 4 


as the king has. performed. — Added to the main! te^t (which is difficult 
because of its abundance of rare words) is a trite lit tie story of how the 

page subsequently accomplished a routine heroic foat in subduing two 
lions. This is evidently a later vulgarisation. • 

Three short texts, probably all post-Sasanian, arc purely practical in 
content. One is a little Pazand work on the xweskdrih i redagdn ("duties 
of boys"), 1 containing simple precepts for schoolboys. Another is a brief 
manual on letter-writing (ndtnag-nibesisnik), 2 ; and the third contains a 
specimen after-dinner speech. 3 Ibn an-Nadim lists other practical man- 
uals on such subjects; as polo and falconry. 4 ; 

rr. Glossaries 

The Sasanians made, it seems, no dictionaries or grammars of their 
language; but there exists a glossary of Avestan words and their Middle 
Persian equivalents (Frahang I oim-mag), 5 based on the Zand, from which 
citations are made. In this there is an attempt at defining grammatical 
categories. A fragment survives of a glossary of Aramaic terms in Pahlavi, 
with their Middle Persian equivalents, 5 which belongs probably to the 
9th or 10th century. A later work of this nature, the Frahang i pahlawig,' 1 
includes some. archaic Iranian words and their commoner synonyms. 
In it the ideograms are often corruptly or carelessly written, and artificial 
forms also occur. 

12. Prose Works of Entertainment 

Among later Sasanian translations of foreign books were prose works 
of entertainment. One was the Hellenistic romance later known as 
Wd-miq wa *Aihr&* said to have been dedicated v in translation to Xosrau 

1 Dakmesteter J A 1880, 355 ff.; Aktia Ptuend Texts, 73 H.; Junker So. Htidelberger AW., 
1912, Abh, 15 j Freiman Dastwr Hoshang Mem. Vol., 482 ff. 

! Jamasp-Asana Pahi. Texts 132-4.0; transcribed and traosL by Za;:hner B$0$ IX 1 (1937), 

3 Jamasp-Asana Pahl. Texts 155-59; Tavadia /. of the K.R, Catna Oriental Inst. XXIX (1935), 

* Fihrist ed. FlOqel, 314.. 21-22, 315. 16. 

6 See West GIF ii, 37 i aJso ed- by C. Salsmanm in Travaux de la jt session du congres inter- 
national des oridhtalisUs, St Petersbourg 1876, II, 403-593; subsequently ed. by H. Reichelt Der 
Frahang i aim (Wien, 1900) and in WZKM XIV (1900), 177-213, aad XV (1901), 117-186. 

4 See Barr BSOS VIII (1936} 391-4.03. 

' Ed. Junker TheFfahang i Pahlavlk (Heidelberg, 1912); see further B. Geiger WZKM XXV 
(1912), 294-306; Schaeder Iranische Beitrage I, 225-54; E. Ebeliko Mitteil. d. AUcrientaliscken 
Ges., Leipzig, XIV i (1943:); Junker Das F-rakang i Paklavik in seiekengetnasser Anordnwitg (Leipzig, 

• See M. Shah Proceedings of the XXIlIrd International Congress 1/ Orientalists (Cambridge, 
J 954li 160-1. 



I. From India came such works as the Tiiti Ndmag, Sindbdd Ndmag, 
Balauhar ud Buddstzf 1 and Kalilag ud Dimnag. The last-named derived 
from the Pancatantva, and according, to its preface 2 (preserved in the 
translation of Ibn Muqaffa') was rendered into Persian by Burzoe, a 
physician of the time of Xosrau. There were also separate short stories 
such as the two in the Letter of Tansar. The Indian tales came evidently 
from more than one source, some perhaps directly from Buddhist com- 
munities in the north-east, 3 others through Manichaean intermediaries. 4 
They are characterised by being told avowedly for some moral purpose, 
or being set in an edifying framework. 

Persia herself had an oral tradition of story-telling. "The professional 
story-teller ... had his place at court, and the richness of his repertoire 
is implied by the fact that he was forbidden ever to repeat himself, 
unless at the king's command." s Probably for a long time the short story 
was considered unworthy of being written down; but, under the stimulus 
perhaps of these foreign works, collections of tales were made. Their 
exact nature is unknown. The most famous is the Haz&r Ajsdn ('Thou- 
sand Tales"), said to be the origin of the "Thousand and One Nights", 
but Ibn an-Nadim says that he saw the Persian book more than once, and 
found it a collection of tedious traditions only. 6 Possibly the dignity 
attaching to the written word led to a certain restraint in the writing of 
indigenous short stories during the Sasanian period. 

13. The MS. -tradition 

The Zoroastrian MS.-tradition is relatively late. A colophon exists 
belonging to a MS. copied in 1020 AX.; but the MS. (K 43) in which it 
is preserved was itself written in the 16th century. The oldest extant 
MSS. belong to the 14th century. 7 The Indian MS.- tradition depends on 
the Iranian one, MSS. having been brought to India from Iran from the 
13th century onwards; 8 but a few works (including the Skand-Gumdmg 
Wizdr) were preserved (in Pazand version) only by the Parsis. There 

«* I*!?' « L V° BS0A , S XX (I957) 389 tf ' and Ue Wisdom Walahvar, a Christian legend of 

the Buddha (Loudon, 1957). * J 

,A S t* N ?*™f ™ BuTZ " i$ E '"'"'*"* ™ *» ^Che Kalila wa Drum*, iibersetti und erl&utert 

(SchTiflen d. Wissenschaftlieken Gesttlstkajt in Slrassburg, 12), 1912. 

3 See Asia Major n.s. V i (1955), 53 ff. 

1 See W. Bang Le Mv-sion XLIV (1.931), i- 3 6; Heknikg BSOAS XI (1045), 465-87 

5 See JRAS (1957), 34. 

« Fihrist ed. FlCgel 304.19; see v. Ham me r-Purcst a ll J A 1839, 171-6. 

1 On colophons, palaeography etc. see particularly Bartholomae Die Zend Hatuisckriften 
unvala Colophons. 

" See West GIF ii 80. 

Handbuch der Oriental is tik, Abt. I, Bd. IV, Abschn. 2, Lfg. r 



seems to have been a break in continuity of the Pahlavi tradition in 
India; 1 but from the 14th century onwards Indian copyists were again 
active, although a number of important works did not reach their country 
till the iBth and 19th centuries. A few trifling onhographic divergences 
characterise the Indian MSS. The standards of copying in both countries 
were in the main low, carelessness aggravating the. difficulties of a dead 
language and ambiguous script. 

The problems presented by both were and are considerable. The 
Manichaean texts provide help, in that they have preserved the Middle 
Persian of the early Sasanian period, in an orthography which gives a 
clear indication of most of the sounds. From them it appears that the 
orthography of the Phi. books represents in the main a pre-Sasanian 
stage of the language, but that syntax and vocabulary are in general 
that of late Middle Persian. Manichaean MP. is the language of Pars. 
The MP. of the Zoroastrian books is a literary koine, merging into Persian, 
with many northern forms mingling with those of the south. Since 
Persian was known to the copyists, there was naturally an unconscious 
temptation for them to modernise syntax and spellings still further, 
especially where the older forms of words had usually been ideogramma- 
tically represented. This unintentional modernisation was joined with 
some deliberate but misguided attempts at archaising, and the two 
processes produced bastard forms which never existed in the spoken 
tongue. Thus the language of the Phi. books, like so much of their 
contents, is a tangled web of ancient and modern, of authentic and spec- 
ulative. It is preserved, moreover, in a script so ambiguous that, for 
instance, such different words as xdnag and dhog look exactly alike, and 
one stroke alone distinguishes wistdxw from gitigan. A misreading, or the 
faulty transposition of an ideogram by a copyist, and the sense becomes 
distorted. Fortunately in this century not only has the Manichaean mate- 
rial greatly enlarged knowledge of Middle Persian, but the discovery 
of more Phi. inscriptions, and of older documents written in a less cursive 
form of the script, has helped to resolve many of the difficulties of the 
MS.-tradition; but much editing and re-editing remains to be done. 2 

1 See Tavadia ZDMG 98 (1944), 303 ft 

* The short VatQa Nash has only recently been edited for the first •:ime by F. M. Kotwal (Bom- 
bay, 1966), who has shown this particular work to be a 18th-century lorgery. Other unedited texts 
are still to be found in the libraries 01 Navsari and Bombay, 

NOTE; This article was sent to press in 1958. Bibliographical seditions to the notes -have 
been made up to April 1967. 




1. Introductory 
Three factors separate the Manichaean texts from other Middle 

Lr bZ LI ^ Palm >— s -p tl familiar to Mani at his birthplace 

2*2 ^ ° r u^ 3ramai1 la " gUa * es - L This scri Pt ^ a clear and 

scribal tradition A second factor is that of content. Although the Ma- 
tt 71 T" ** Zoroastrian ones ' contai * so ™ &*%* ™«£ 

the proportion is in their case minute. In the main thev are strictly 
religious; and the Manichaean religion, although preached by a prophet 
of Iranian blood, had Semitic and Graeco-Senntic elements which give 
for™ reSPGCtS ' f ° rei§n CharaCt£r ' both in ***** -d liteLy 

The Manichaean writings were first discovered in the early decades of 

oaL'Tch^T 1 : Sand t ned minS ° f m ° naSteries m ^ ^rf an 
IT A i ; T* TurkeStan " 2 The MSS- were of various kinds-the 

leather ^' < ^f^ "« ^^ *** ** were made of 

often beautifully illuminated. Part of one tiny MS. survives almost 

Tt~ ( 8 b fr i bUt f heiWiSe aU the MSS " ha« been reduced 
to fragments, probably largely through the Z eal of Muslim conquerors, 

I I? + H " K ™ & »"«»«* I, IV 1 73 , with icfcrcnoes. 

Most of the material was discovered by 4 German P,~H ift ™ „ . . - 

Institut fur OrieatfoKchun*, Berlin: and about aTurV™" ^L I?*l P * rt ° f " » EOW ia the 
und der Literatur, Mainz A small nnmb* <$t,JZ^ I- '" *? Akademie *** Wwenscbrftan 
Leningrad; and afew others Te Zd b JapS e iSSE? * ^ »**«**»* * •» 
fragment in the Bibliothequ* Native, Parish H^o SS^S °^ ^ * "V" 1 " 
some half-dozen in the British Museum. The ?A„ S k ? 5f? 1 ?/* Ud f l^l"- 5 ""* 
expeditions have been catalogued in A Ca^l^r^.r ?;.!. OTpt iound b ^ the German